@article{akahomeje20261, title = {Bridging creativity and structure: how generative AI aligns innovation and project management practices in agricultural initiatives}, author = {Akahome J.E.}, year = {2026}, journal = {International Journal of Managing Projects in Business}, pages = {1-21}, doi = {10.1108/IJMPB-10-2025-0456}, abstract = {Purpose – This study explores how generative artificial intelligence (AI) bridges creativity and structured project management in agricultural initiatives. It investigates how AI-supported project management enhances innovation, collaboration, and adaptive decision-making within the Nigerian agricultural sector. Design/methodology/approach – The research adopts a qualitative multiple-case study approach, drawing information from 25 agricultural cooperatives, extension services, and AI-enabled farm projects across Northern and Southwestern Nigeria. Semi-structured interviews and document reviews were analyzed thematically using Dynamic Capabilities Theory and Socio-Technical Systems Theory as guiding lenses. Findings – The study reveals that generative AI transforms project management roles from administrative supervision to facilitation of creative decision-making. AI platforms foster fast collaboration, simulation, and experimentation and allow teams to test innovative ideas without jeopardizing structured workflows. Primary enablers include leadership support, strategic alignment, training, and reliable digital infrastructure. However, barriers such as limited digital literacy, resistance to change, and ethical concerns constrain adoption. The findings led to the development of the Techno-Creative Alignment Framework, which explains how human collaboration, ethical awareness, and AI intelligence interact to balance innovation and structured project execution. Originality/value – This research contributes to the limited scholarship on AI's double role in fostering creativity and structure within project-based agricultural innovation, offering both conceptual and contextual insights relevant to developing economies. Beyond national relevance, it offers international contributions by positioning developing economies as laboratories of adaptive innovation to illustrate how AI can serve as a transformative equalizer instead of just a tool for efficiency. The findings provide a comparative perspective for global project management scholarship to highlight how lessons from the Global South can inform responsible AI integration in agricultural innovation worldwide. © Emerald Publishing Limited all right reseved.}, keywords = {Agricultural projects; Dynamic capabilities theory; Generative artificial intelligence; Innovation management; Project management; Socio-technical systems theory}, } @article{kikism20262, title = {AI on the path to good decisions}, author = {Kiškis M.}, year = {2026}, journal = {Sustainable Futures}, volume = {11}, pages = {101644}, doi = {10.1016/j.sftr.2026.101644}, abstract = {This article examines objections to the use of artificial intelligence (AI) in decisions affecting humans and argues that most objections rest on anthropocentric assumptions rather than evidence about decision quality. Critics often presume that human decision makers are uniquely capable of moral and contextual judgment, while AI systems are inferior, opaque or hostile. The article challenges this view arguing that modern AI systems are an expression of collective human intelligence and ethics, built on the image of human cognition, trained on curated human decisions and operating on best decision making frameworks. Drawing on empirical studies the article shows that modern AI systems offer unique advantages in decision-making, thus may already in some domains be as good at making good decisions as an average individual human. The article deconstructs main criticisms of AI decision-making and introduces the novel argument of inseparability between human and AI decisions. Human and AI contributions are increasingly intertwined, AI involvement is latent and appropriated by humans, making existing accountability frameworks based on a clear human–AI boundary obsolete. The article advocates for the development of agnostic decision-making frameworks that apply universal accountability to both human and non-human agents and provide a path to better decisions. © 2026}, keywords = {Accountability; Ai; Artificial intelligence; Collective intelligence; Decision-making; Ethical AI; Good decisions}, } @article{xumzhangwguptashuangj20263, title = {The art of the successful dance: The investigation of AI adoption on firm performance}, author = {Xu M. and Zhang W. and Gupta S. and Huang J.}, year = {2026}, journal = {Journal of Strategy and Innovation}, volume = {37}, number = {2}, pages = {200576}, doi = {10.1016/j.jsinno.2026.200576}, abstract = {With the global technological developments in the past decades, AI has been widely adopted in firms' business and marketing strategies, aiming to create more positive outcomes. While most of the extant literature studying a firm's performance outcomes focuses on a single factor (e.g., salespeople and supply chain productivity), there is a lack of research examining the AI's impacts on the total changes in a firm's production efficiency from multi-factor inputs and the changes in the overall customer base, instead of individual customers' experience. Accordingly, this paper aims to understand the impacts of AI adoption on the overall business performance from the perspectives of firms, employees, and customers, and explore the moderating effect of a firm's market power and external AI policy. It conducts interdisciplinary research, using a mixed-method approach, with a combination of 2852 Chinese listed firms' data spanning from 2010 to 2022 and semi-structured interviews with 5 board-level directors in Chinese listed firms. The results reveal the impacts of AI adoption on the increase of total firm productivity, positive change in employment, and reduced customer concentration. The moderation effects of a firm's market power and AI policy are confirmed for these impacts. It encourages future studies to investigate AI adoption in different business environments from the firm, employee, and customer perspectives. © 2026}, keywords = {AI adoption; Artificial intelligence; Firm performance; Market power; Social policy; Commerce; Personnel; Sales; AI adoption; Business strategy; Firm Performance; Market Power; Marketing strategy; Performance outcome; Production efficiency; Salespeoples; Social policy; Technological development; Artificial intelligence}, } @article{patrasp20264, title = {Mapping the causal structure of AI governance factors influencing quality assurance under Industry 4.0 regulations}, author = {Patra S.P.}, year = {2026}, journal = {TQM Journal}, pages = {1-21}, doi = {10.1108/TQM-01-2026-0029}, abstract = {Purpose – The increasing deployment of artificial intelligence (AI) in Industry 4.0 has revealed limitations in traditional quality assurance systems, which are not designed to govern algorithmic decision-making and autonomous operations. This study aims to empirically examine how AI governance dimensions causally influence quality assurance outcomes in AI-enabled industrial systems. Design/methodology/approach – A Delphi-based expert elicitation involving 10 specialists from industry, technology, policy and regulation was conducted. The fuzzy decision-making trial and evaluation laboratory (fuzzy DEMATEL) method was applied using linguistic variables and triangular fuzzy numbers to model cause-and-effect relationships among governance dimensions. Findings – The analysis identifies a clear causal hierarchy. Algorithmic transparency and cybersecurity assurance emerge as dominant causal drivers, supported by data governance quality, continuous improvement culture and integration of ISO and AI standards. Ethical accountability, regulatory compliance readiness, stakeholder trust, process reliability and top management commitment appear as effect dimensions. Strong feedback loops indicate non-linear governance maturity and cascading effects. Research limitations/implications – This study empirically maps the causal relationships between AI governance and quality assurance using fuzzy DEMATEL; however, several limitations should be acknowledged. First, the expert panel comprised 10 professionals primarily from highly regulated sectors in developed economies, which may limit generalizability to SMEs, emerging economies or less-regulated industries. Second, the fuzzy DEMATEL approach relies on expert judgements expressed through linguistic variables, and residual subjectivity may persist despite Delphi refinement. Third, the cross-sectional design does not capture the dynamic evolution of governance-quality relationships over time. Finally, the threshold-based filtering emphasizes dominant relationships and may overlook weaker yet meaningful interactions, while implementation contingencies across organizational contexts remain unexamined. Originality/value – The study offers one of the first empirical causal mappings linking AI governance and quality assurance. By integrating total quality management principles with AI governance, it reframes governance as a systemic quality capability and provides actionable guidance on sequencing governance investments. © Emerald Publishing Limited}, keywords = {AI governance; Causal hierarchy; Fuzzy DEMATEL; Industry 4.0; Organizational governance; Quality assurance; Artificial intelligence; Decision making; Ethical aspects; Fuzzy set theory; Industrial economics; Mapping; Regulatory compliance; Algorithmics; Artificial intelligence governance; Causal hierarchy; DEMATEL; Fuzzy Decision making; Fuzzy DEMATEL; Linguistic variable; Organisational; Organizational governance; Quality assurance systems; Quality assurance}, } @article{tranletuyettnguyenkm20265, title = {Responsible AI and career sustainability: the intersectional role of knowledge, emotion, and capability in Vietnam}, author = {Tran Le Tuyet T. and Nguyen K.M.}, year = {2026}, journal = {Cognition, Technology and Work}, volume = {28}, number = {2}, pages = {501-532}, doi = {10.1007/s10111-025-00851-4}, abstract = {This study investigates how responsible AI signals (RAS) such as autonomy, justice, beneficence, explainability, and nonmaleficence enhance employees’ career sustainability, with particular focus on dynamic capability, AI emotional response, and AI knowledge management (knowledge sharing, acquisition, and application). Grounded in the Cognition–Affect–Conation (CAC) framework, the study extends its scope from psychology to human resource management by explaining how cognitive, emotional, and behavioral mechanisms jointly shape employees’ responses to responsible AI practices. A quantitative research design was employed using an online questionnaire, gathering responses from a sample of 717 employees in Vietnam and analyzed using PLS-SEM. The findings reveal that RAS strongly promotes AI emotional response, dynamic capability, and knowledge management processes, including knowledge sharing, acquisition, and application. In turn, AI emotional response, dynamic capability, and the application and sharing of knowledge exert significant positive effects on employee well-being and innovation performance, whereas knowledge acquisition shows no meaningful impact. The study advances theory by integrating the CAC framework with Responsible AI principles to explain how employees adapt and collaborate with AI in organizations. Practically, the findings indicate that adopting responsible AI principles can enhance employee creativity, emotional engagement, and adaptability by promoting knowledge sharing, supportive policies, and transparent AI practices. Managers are encouraged to design learning-oriented environments, continuous AI ethics training, and participatory mechanisms that allow employees to engage with AI fairly and autonomously, thereby fostering well-being, innovation, and long-term career sustainability. © The Author(s), under exclusive licence to Springer-Verlag London Ltd., part of Springer Nature 2026.}, keywords = {AI knowledge management; Career sustainability; Dynamic capability; Emotional response; Responsible AI; Behavioral research; Employment; Human resource management; Knowledge acquisition; Knowledge management; Knowledge transfer; Mergers and acquisitions; Professional aspects; Sustainable development; AI knowledge management; Career sustainability; Dynamics capability; Emotional response; Knowledge application; Knowledge-sharing; ON dynamics; Responsible AI; Viet Nam; Well being; Personnel training}, } @article{omarbhon20266, title = {An analytical framework for business leaders to assess SMEs’ readiness for AI adoption}, author = {Omar B. and Ho N.}, year = {2026}, journal = {Strategy and Leadership}, pages = {1-25}, doi = {10.1108/SL-03-2026-0118}, abstract = {Purpose – This study introduces a practical analytical framework that enables Small and Medium-sized Enterprises (SMEs) leaders to assess their business’s readiness to adopt Artificial Intelligence (AI) technologies across key business dimensions, prioritise resource allocation and manage the adoption process effectively. Design/methodology/approach – The authors employed an integrative review of contemporary AI Readiness literature to synthesise existing theoretical perspectives into a six-dimensional AI Readiness Analytical Framework, including: Business Strategy, Resource Orchestration, Skills and Knowledge, Ethical Practice, Culture and Data Management, followed by direct validation and feedback from 17 business leaders, primarily from SMEs. Findings – The proposed AI Readiness Analytical Framework is validated as both theoretically robust and practically relevant, with all dimensions rated highly in terms of importance by the SME leaders. Strategy is ranked as the most important dimension, highlighting the need for clear vision, leadership, defined use cases and effective change management. Resource Orchestration is ranked lowest, reflecting SMEs’ preference for accessible, low-cost AI tools; however, resource requirements are expected to increase as AI adoption matures. Originality/value – This study introduces a leadership-centred AI Readiness Analytical Framework tailored to SMEs that addresses gaps in large-enterprise models. Its originality lies in both the validation of the framework by SME leaders and the positioning of Ethical Practice as a standalone dimension. The framework functions as both a diagnostic and strategic tool, enabling proactive, capability-aligned AI adoption. © 2026 Emerald Publishing Limited}, keywords = {AI adoption; AI ethics; AI Readiness; Leadership; SME; Strategy}, } @article{buinskienrmiceikienahernndeztamurejosaurajr20267, title = {The knowledge of ethical AI decision-making: A behavioral economics perspective}, author = {Bužinskienė R. and Miceikienė A. and Hernández-Tamurejo Á. and Saura J.R.}, year = {2026}, journal = {Journal of Innovation and Knowledge}, volume = {14}, pages = {100967}, doi = {10.1016/j.jik.2026.100967}, abstract = {Focusing on artificial intelligence (AI) ethics and behavioral economics, this paper demonstrates how integrating behavioral insights can inform ethical guidelines for AI-based systems. Recent developments conceptualize AI as a key to critical sectors such as finance, marketing, and healthcare, thereby prompting a widespread recognition of its potential to reinforce or mitigate societal biases. However, while AI ethics has emphasized technical concerns such as data fairness and algorithmic transparency, insights from behavioral economics remain largely untapped. To bridge this gap in the literature, the present study uses a three-stage methodology. First, a systematic literature review (SLR) is conducted to identify gaps in AI ethics frameworks and to establish how key behavioral economic principles can enhance transparent decision making. Second, in-depth interviews are performed to collect additional insights and to compare how the results compare with those obtained from the SLR. Third, a topic modeling-based model, latent Dirichlet allocation (LDA), supported by the Computer-Assisted Text Analysis (CATA) framework, is applied to the collected interview corpus in order to identify the main topics in the data. The results reveal that ethical AI is increasingly understood by professionals not only as a technical issue but also as a behavioral and organizational one. Specifically, we identify five main topics from the interview content, namely Data & transparency, Behavioral insights & design, Legal/ethical frameworks, Implementation challenges & infrastructure, and Civic/stakeholder engagement. The first three of these topics dominate in the current discourse on AI decision making. The results also reveal that cognitive biases, such as anchoring and confirmation bias, affect the data that feed AI algorithms and user interaction with AI-driven outputs. If these biases remain unaddressed, they can amplify ethical risks. However, awareness of such patterns can inform practical design and governance strategies so as to improve transparency, fairness, and user engagement. Finally, based on the findings, four research proposals are advanced to clarify how behavioral economics can be systematically integrated into ethical AI decision making. Copyright © 2026. Published by Elsevier B.V.}, keywords = {AI; Artificial Intelligence; Behavioral Economics; Decision-making; Ethics; Knowledge}, } @article{brandafciccozzim20268, title = {The comfort of automation: why cognitive sovereignty matters in AI-driven life sciences}, author = {Branda F. and Ciccozzi M.}, year = {2026}, journal = {Artificial Intelligence in the Life Sciences}, volume = {9}, pages = {100158}, doi = {10.1016/j.ailsci.2026.100158}, abstract = {The integration of artificial intelligence (AI) into the life sciences is radically transforming research, clinical diagnosis, and therapeutic development processes, redefining the relationship between knowledge, decision-making, and responsibility. Advanced tools, from generative models to clinical assistants such as ChatGPT Health, offer greater efficiency, predictive power, and access to data, but carry significant risks of automation bias, epistemic delegation, and loss of professional skills. This article analyzes how the extensive use of AI can threaten cognitive sovereignty, i.e., the ability of researchers and professionals to critically evaluate and contextualize information generated by algorithms. It examines the emerging regulatory landscape, with a focus on the EU Artificial Intelligence Act, Food and Drug Administration (FDA) guidelines, European Medicines Agency (EMA) Good Machine Learning Practice (GMLP) principles, and World Health Organization (WHO) recommendations, which aim to ensure human oversight, transparency, and accountability. Technological tools and training approaches are discussed to mitigate risks such as silent errors, algorithmic dependence, and skill deterioration, promoting AI integration that reinforces human judgment without replacing it. The analysis highlights that the future of life sciences will depend not only on the technical capabilities of models, but also on the critical awareness with which they are used, focusing on training, governance, and responsible AI design. © 2026}, keywords = {Artificial intelligence; Automation bias; Cognitive sovereignty; Human oversight; act; Article; artificial intelligence; automation; automation bias; cognitive sovereignty; critical thinking; decision making; decision support system; epistemic delegation; EU Artificial Intelligence Act; European Medicines Agency; Food and Drug Administration; good clinical practice; Good Machine Learning Practice; human; information processing; practice guideline; social responsibility; World Health Organization}, } @article{senarakc20269, title = {Cyber resilience skills for AI-enabled incident response in port operations}, author = {Senarak C.}, year = {2026}, journal = {Transportation Research Interdisciplinary Perspectives}, volume = {37}, pages = {102033}, doi = {10.1016/j.trip.2026.102033}, abstract = {Ports are increasingly deploying artificial intelligence (AI) to support cybersecurity monitoring, analysis, and automated response. While prior research has focused on the technical performance of AI-enabled response systems, far less attention has been given to the human skills required to govern cyber incident response during active disruption. In port environments—where cyber incidents can immediately affect safety, cargo flows, and service continuity—response effectiveness depends not only on automation, but on managerial judgment exercised under time pressure and uncertainty. This study reframes cyber incident response capability as a set of decision-oriented skills enacted by port operations managers during the Respond phase. Using a two-round Delphi method with experts from port operations, cybersecurity, and governance fields in Thailand, the study identifies and validates six core cyber resilience skill domains: Operational Impact Interpretation; Response Coordination and Order-of-Operations Leadership; Criteria-Based Reporting and Stakeholder Communication; Containment–Continuity Trade-off Management; Human–AI Supervision and Decision Governance; and Learning and Response Improvement. The findings advance cyber resilience and port management literature by making the human governance layer of AI-enabled response analytically explicit. The resulting skill framework provides a foundation for future empirical research and offers practical guidance for training, role design, and policy development in AI-enabled port cybersecurity governance. © 2026 The Author(s).}, keywords = {AI-enabled cyber incident response; Cyber resilience; Delphi method; Human–AI governance; Port operations management}, } @article{yigitcanlartdavidamarasinghersenadheerashossaintyextaeihagha202610, title = {Governing Urban AI from the Frontline: A Stage-Gate Framework for Municipal Algorithmic Decision-Making}, author = {Yigitcanlar T. and David A. and Marasinghe R. and Senadheera S. and Hossain T. and Ye X. and Taeihagh A.}, year = {2026}, journal = {Smart Cities}, volume = {9}, number = {5}, pages = {81}, doi = {10.3390/smartcities9050081}, abstract = {Artificial intelligence (AI) is increasingly embedded in how cities are governed, shaping decisions on mobility, land use, public services, and environmental management. Yet urban AI is predominantly governed through fragmented frameworks designed at national or corporate scales, offering limited guidance for municipal decision-making and overlooking place-specific social and ecological consequences. As the level of government closest to everyday urban life, cities are uniquely positioned to steer AI toward public value, but face persistent tensions between efficiency, equity, accountability, and sustainability. This paper argues that responsible urban AI cannot be governed through top-down or one-size-fits-all approaches. To address this, the study aims to conceptualise and advance a ground-up model of responsible urban AI governance that places cities and local governments at the centre of decision-making. It addresses the following research question: How can municipal authorities translate high-level ethical principles into practical, context-sensitive governance arrangements that respond to local capacities, risks, and public values? Drawing on global governance principles and illustrative city experiences, we propose a locally grounded, stage-based framework for municipal AI governance. The framework addresses institutional capacity gaps, fragmented responsibilities, and algorithmic externalities, advancing a participatory, place-sensitive, and adaptive model that aligns urban AI innovation with democratic legitimacy, social justice, and sustainable urban futures. © 2026 by the authors.}, keywords = {AI governance; algorithmic decision-making; artificial intelligence (AI); city governance; local government AI; municipal AI; municipal governance; participatory oversight; responsible AI; urban AI}, } @article{zaidanetrubyjibrahimiahoppet202611, title = {Hybrid global governance for responsible and inclusive Artificial Intelligence: Proposing a new Sustainable Development Goal 18}, author = {Zaidan E. and Truby J. and Ibrahim I.A. and Hoppe T.}, year = {2026}, journal = {Technology in Society}, volume = {85}, pages = {103159}, doi = {10.1016/j.techsoc.2025.103159}, abstract = {Artificial Intelligence (AI) is increasingly recognised as a transformative force in advancing and potentially undermining, the United Nations Sustainable Development Goals (SDGs). While AI can drive innovation to benefit SDG's , it also amplifies risks such as bias, surveillance abuse, and environmental impacts. Existing governance approaches remain fragmented, with principles-based, rights-based, risk-based, and ethics-based frameworks operating in silos. This paper proposes integrating them into a hybrid governance framework by introducing a new SDG 18 on Responsible and Inclusive AI for Sustainable Development. The proposed SDG 18 can offer a politically feasible, soft-law mechanism to align AI innovation with SDGs of societal well-being, global equity and sustainability. Designed as a pervasive Goal across the SDGs, the proposal would be beneficial by embedding human-centric values, environmental stewardship, and collaborative governance into AI oversight, and seek to advance the achievement of the remaining SDGs with AI tools. The framework contributes to debates on international technology governance by providing policymakers with an adaptable tool for managing AI's cross-border impacts, promoting trust and ensuring that technological progress serves public interest. © 2025 The Authors.}, keywords = {AI governance; Digital inclusion; Ethical AI; Hybrid governance framework; International cooperation; Responsible AI; SDG 18; Technology policy; Artificial intelligence; Environmental impact; Ethical technology; Public policy; Sustainable development; Sustainable development goals; Artificial intelligence governance; Digital inclusion; Ethical artificial intelligence; Global governance; Hybrid governance framework; Responsible artificial intelligence; Risk-based; Sustainable development goal 18; Technology policy; United Nations; artificial intelligence; governance approach; information technology; sustainability; Sustainable Development Goal; technology policy; United Nations; International cooperation}, } @article{quczhaoxxiex202612, title = {Dynamics of social and cultural institutions in AI governing: a PMC study based on China’s AI policy}, author = {Qu C. and Zhao X. and Xie X.}, year = {2026}, journal = {Data and Policy}, volume = {8}, pages = {e24}, doi = {10.1017/dap.2026.10066}, abstract = {This study focuses on a multidimensional evaluation of China’s artificial intelligence (AI) governance policies by employing the Policy Modeling Consistency (PMC) index model to systematically analyze the performance and coherence of these policies in terms of goal setting, technological attributes, risk governance, and socio-cultural institutional embedding. The study conceptualizes AI as a novel form of cultural and social technology rather than merely an intelligent agent. This perspective highlights AI’s role in reorganizing and restructuring information flow, cultural expression, and institutional operations. By constructing a PMC evaluation framework that integrates technological, risk, and socio-cultural dimensions, this research reveals both strengths and weaknesses in the current policy system, particularly significant gaps in embedding socio-cultural institutions. Combining theoretical and empirical analyses, the PMC model offers an effective framework for understanding the institutional dynamics of AI governance in China. The findings enrich theoretical perspectives on AI governance and provide empirical support for policy formulation. © The Author(s), 2026. Published by Cambridge University Press.}, keywords = {AI governance; artificial intelligence policy; policy modeling consistency; socio-cultural institutions; socio-cultural technology}, } @article{joshiubaisilskinibsmehtamdattas202613, title = {AI ethics in Indian healthcare: a scoping review of national and international guidelines on privacy, data protection, and security}, author = {Joshi U. and Baisil S. and Kini B S. and Mehta M. and Datta S.}, year = {2026}, journal = {BMC Medical Ethics}, volume = {27}, number = {1}, pages = {86}, doi = {10.1186/s12910-026-01435-1}, abstract = {Objective: To systematically map and critically analyze the provisions within Indian national frameworks (Digital Personal Data Protection Act 2023, ICMR guidelines) and key international ethical AI guidelines that address privacy and data security for AI-driven diagnostic tools, specifically within the Indian public health context. Methods: A scoping review was conducted following the PRISMA-ScR framework. A comprehensive search of government publications, intergovernmental organization reports, and academic literature was performed to identify relevant national and international guidelines. Data were charted and synthesized thematically. Results: The analysis reveals a fragmented Indian governance landscape, characterized by a principles-based national AI strategy (#AIforAll), sector-specific ethical guidelines (ICMR), and a new, general-purpose data protection law (DPDP Act). While the DPDP Act establishes foundational data fiduciary obligations, its broad exemptions for public health and research create ambiguities for data-intensive AI. International frameworks serve as comparative reference points; the EU AI Act, in particular, offers a contrasting granular, risk-based regulatory model. Key tensions emerge between the universalist principles of international guidelines and the need for a “situated ethics” approach that addresses India’s unique challenges, including the digital divide, data quality issues, and the difficulty of operationalizing meaningful consent. Conclusion: Existing guidelines provide a foundational but incomplete framework for the safe and ethical deployment of AI diagnostics in Indian public health. There is a critical need to bridge the gap between high-level principles and on-the-ground implementation. This requires developing sector-specific regulations under the DPDP Act, establishing robust standards for data and algorithmic audits, and fostering a cooperative federalism approach that harmonizes international standards with the socio-technical and constitutional realities of India. © The Author(s) 2026.}, keywords = {Artificial Intelligence; Data Privacy; Healthcare Governance; Medical Ethics; Public Health Policy; Scoping Review; article; artificial intelligence; clinical practice guideline; data privacy; data protection; ethics; health care; health care policy; human; medical ethics; privacy; public health; scoping review; security; systematic review}, } @article{penatcunhaascordeirojvvictorinog202614, title = {Enhancing corporate innovation through generative AI: A case study on Human–AI collaboration}, author = {Pena T. and Cunha A.S. and Cordeiro J.V. and Victorino G.}, year = {2026}, journal = {Social Sciences and Humanities Open}, volume = {13}, pages = {102867}, doi = {10.1016/j.ssaho.2026.102867}, abstract = {Although Generative Artificial Intelligence (GenAI) is revolutionizing corporate innovation processes, little is known about how it fits into Design Thinking (DT). This study uses an exploratory case study in a corporate innovation workshop to examine how GenAI affects three essential DT components: trend identification, brainstorming, and feedback. Using a mixed-methods approach, we assessed participants' perceptions before and after DT activities with and without GenAI help. Our findings demonstrate that GenAI accelerates ideation, improves creative confidence, and promotes trend analysis and feedback integration. However, ethical concerns regarding AI bias and transparency also exist, which emphasizes how important it is to use AI ethically. Human judgement is necessary at every evaluation stage to ensure morally sound and contextually relevant decision-making, even while GenAI promotes efficiency and alternative thinking. This study contributes to the emerging discourse on AI-human collaboration in innovation, offering insights for organizations seeking to integrate GenAI into human-centered design processes. © 2026 The Authors.}, keywords = {Artificial Intelligence (AI); Corporate innovation; Design Thinking (DT); Ethics; Generative Artificial Intelligence (GenAI)}, } @article{nguyentpl202615, title = {From return on investment to return on health: the CARE framework for AI governance, accountability, and equitable transformation in healthcare systems}, author = {Nguyen T.P.L.}, year = {2026}, journal = {International Journal of Data Science and Analytics}, volume = {22}, number = {1}, pages = {184}, doi = {10.1007/s41060-026-01152-3}, abstract = {Pharmaceutical marketing analytics effectiveness has traditionally been evaluated using financial return on investment (ROI), with limited integration of patient-centered health outcomes into optimization objectives. This study introduces the CARE framework—causal intelligence, accountability analytics, responsible engagement, and equity outcomes—as a structured analytical architecture for reframing marketing evaluation around return on health (ROH). Rather than considering health impact as a secondary consequence of revenue optimization, CARE integrates causal inference, transparency mechanisms, and fairness constraints directly within the optimization objective. The study employs a simulation-based validation design using synthetic datasets parameterized with publicly reported healthcare adherence, access, and affordability distributions. The analytical architecture integrates marketing mix modeling (MMM), multi-touch attribution (MTA), and incrementality testing within a governance-constrained causal estimation structure. In the simulated environment, patients were assigned to treatment and control groups to enable causal effect estimation. Double machine learning techniques were used to isolate the independent effect of marketing exposure while controlling demographic, socioeconomic, and clinical confounding factors. The optimization procedure also incorporated fairness constraints to prevent disproportionate allocation of resources across population subgroups. Simulation analyses examine how redefining the optimization objective from ROI to ROH changes resource allocation and targeting strategies. Within the modeled environment, health-oriented objective functions shift investment across patient segments and therapeutic areas relative to revenue-maximizing benchmarks, generating comparatively higher projected adherence, therapy initiation, and equity-adjusted outcome indicators. These results reflect model-based analytical demonstrations rather than findings from real-world deployment. The contribution of this study is conceptual and methodological. CARE formalizes a reproducible framework for integrating causal measurement, algorithmic accountability, and equity constraints into healthcare marketing analytics, providing a structured foundation for future for more accountable and health-focused decision making in data-driven marketing systems. © The Author(s), under exclusive licence to Springer Nature Switzerland AG 2026.}, keywords = {Accountability analytics; AI governance; AI-driven optimization; CARE framework; Health equity; Responsible AI; Return on health (ROH); Artificial intelligence; Benchmarking; Commerce; Investments; Marketing; Optimization; Patient treatment; Population statistics; Profitability; Accountability analytic; AI governance; AI-driven optimization; CARE framework; Health equity; Optimisations; Re-turn-on; Responsible AI; Return on health; Returns on investment; Decision making}, } @article{lnenickamclarinvalanikiforovaarudmarkdluterekmsymeonidisdrodrguezbolvarmp202616, title = {Artificial intelligence in policymaking: Mapping integration gaps across the public policy cycle}, author = {Lnenicka M. and Clarinval A. and Nikiforova A. and Rudmark D. and Luterek M. and Symeonidis D. and Rodríguez Bolívar M.P.}, year = {2026}, journal = {Government Information Quarterly}, volume = {43}, number = {2}, pages = {102138}, doi = {10.1016/j.giq.2026.102138}, abstract = {Governments worldwide are increasingly exploring the use of Artificial Intelligence (AI) to support and transform policymaking. Despite the growing number of national AI strategies and initiatives, systematic evidence on how AI is integrated across the stages of the policymaking process remains limited. This lack of clarity risks fragmented integration and slows the diffusion of effective practices. This study addresses this gap by examining how AI-enabled policy actions are incorporated into policymaking processes across eight European countries. We analyze policy documents to identify AI-enabled policy actions and map them across the phases of the public policy cycle. The analysis reveals 33 AI-enabled policy actions and reveals uneven integration across policy stages and countries. While several actions focus on data infrastructure, digital public services, and agenda setting, considerably fewer initiatives target implementation monitoring, evaluation, and responsibility-related governance mechanisms. The study contributes to the literature by (1) systematically mapping AI-enabled policy actions across the policy cycle, (2) identifying integration gaps in current policymaking practices, and (3) outlining directions for future research. Copyright © 2024. Published by Elsevier Inc.}, keywords = {AI; AI governance; Artificial intelligence; Digital government; Policy analysis; Policymaking; Public policy; Public policy cycle}, } @article{pfaudjunga202617, title = {Engineering Trustworthy AI: A Developer Guide for Empirical Risk Minimization}, author = {Pfau D. and Jung A.}, year = {2026}, journal = {IEEE Transactions on Artificial Intelligence}, volume = {7}, number = {5}, pages = {2560-2576}, doi = {10.1109/TAI.2025.3617936}, abstract = {AI systems are increasingly used for critical decisions that transcend all important fields of private and public life. These systems often use empirical risk minimization (ERM) to train powerful prediction models such as deep neural networks. The output of the predictive model runs the risk of unintentional bias, opacity, and other adverse effects. To ensure the safety of these systems, it is vital to consider these risks already in the design stage of the model. The EU acknowledged the potential sensitivity of the predictions/ decisions made about persons, which led to the formulation of the Ethics Guidelines for Trustworthy AI laying down seven key requirements for trustworthy AI. So far, the design of ERM-based methods prioritises accuracy over trustworthiness. This article discusses how key requirements for trustworthy AI can be translated into design choices for ERM components. We map the design space of ML systems to the core objectives of trustworthy AI: fairness, privacy, robustness, and explainability. Our theory is instrumental in the design of trustworthy ML systems that minimize privacy leakage and are robust against (intentional) perturbations during their operation, such as disseminating fake news. The operation of trustworthy ML systems should also be transparent or explainable to its users. Finally, ML systems must be fair and not discriminate against specific user groups. There is an urgent need for a more holistic approach to ML that includes key requirements for trustworthy AI. © 2020 IEEE.}, keywords = {AI ethics; empirical risk minimization (ERM); responsible AI design; trustworthy AI; Prediction models; Risk management; Adverse effect; AI ethic; AI systems; Design stage; Empirical risk minimization; Neural-networks; Prediction modelling; Predictive models; Responsible ai design; Trustworthy AI; Artificial intelligence}, } @article{may202618, title = {Impact of AI pilot zones on corporate ESG performance: Evidence from China}, author = {Ma Y.}, year = {2026}, journal = {International Review of Economics and Finance}, volume = {108}, pages = {105199}, doi = {10.1016/j.iref.2026.105199}, abstract = {This study examines the impact of place-based artificial intelligence (AI) policies on corporate environmental, social, and governance (ESG) performance. Using the staggered rollout of China's National New Generation AI Pilot Zones as a quasi-natural experiment, we employ a difference-in-differences design on A-share listed firms from 2012 to 2023. Our results show that the establishment of AI Pilot Zones significantly improves firms' ESG scores. Mechanism analysis reveals that this policy effect operates through three primary channels: (i) a financial-support channel characterized by increased government subsidies and reduced tax burdens; (ii) a technological-empowerment channel that enhances data utilization and smart investment; and (iii) an organizational-empowerment channel that fosters green managerial innovation and increases the stock of high-skilled human capital. Heterogeneity tests indicate that the ESG-enhancing effect is more pronounced among state-owned enterprises, firms with green investors, and those located in superior business environments, while a “ceiling effect” is observed for firms already certified for green manufacturing. Furthermore, we document significant spatial spillover effects, where ESG improvements diffuse to non-pilot cities through inter-city economic and trade linkages. These findings suggest that AI-related place-based policies function as vital institutional catalysts for corporate sustainability. © 2026 The Author.}, keywords = {AI pilot zones; Data utilization; ESG performance; Green managerial innovation; Smart investment; Spillover effects}, } @article{tuncas202619, title = {The innovation–compliance–perception framework as a lens for AI governance — NLP evidence from Meta's smart glasses and GDPR discourse}, author = {Tunca S.}, year = {2026}, journal = {International Journal of Information Management Data Insights}, volume = {6}, number = {1}, pages = {100408}, doi = {10.1016/j.jjimei.2026.100408}, abstract = {This study examines the interplay between technological innovation, regulatory compliance, and public perception surrounding Meta's AI-powered smart glasses within the EU's GDPR framework. Applying NLP methods—including VADER sentiment analysis, thematic mapping, and n-gram analysis—to a multi-source corpus of 334 articles from eight English-language outlets, the research reveals that positive sentiment clusters around product innovation while significant negative sentiment is tied to GDPR compliance challenges, a pattern statistically confirmed across outlets through chi-square testing (p = 0.0119). Thematic co-occurrence analysis identifies strong intersections between technology and privacy discourse, reflecting persistent concerns about data collection and surveillance. A comparative analysis of EU and UK GDPR frameworks highlights how post-Brexit regulatory divergence under the Data (Use and Access) Act 2025 adds compliance complexity for AI wearables. To interpret these dynamics, the paper develops an Innovation–Compliance–Perception (ICP) framework, demonstrating how governance simultaneously constrains and stimulates technological advancement. Copyright © 2026. Published by Elsevier Ltd.}, keywords = {Artificial intelligence; EU regulations; Natural language processing; Smart glasses; Virtual technologies}, } @article{bhattit202620, title = {Algorithmic Decision-Making and Human Autonomy in Finance: A Systematic Review of AI Ethics Research}, author = {Bhatti T.}, year = {2026}, journal = {Journal of Statistical Theory and Applications}, volume = {25}, number = {1}, pages = {22}, doi = {10.1007/s44199-026-00175-w}, abstract = {Artificial intelligence is rapidly transforming financial decision-making across banking, lending, insurance, auditing, fraud detection, and customer-facing financial services. At the same time, its growing use has intensified ethical concerns related to fairness, accountability, transparency, privacy, trust, and human oversight. Although research on artificial intelligence (AI) in finance has expanded considerably, the literature remains fragmented across disciplinary and application-specific streams, limiting a consolidated understanding of its intellectual foundations and thematic development. This study provides a bibliometric and thematic review of research at the intersection of artificial intelligence, finance, and ethics. Drawing on Scopus-indexed journal articles and a PRISMA-guided screening process, a final sample of 338 articles published between 2000 and 2025 was analyzed using the bibliometrix package in R. The findings show that the field is young but rapidly expanding, particularly after 2019, with strong momentum in recent years. Intellectual structure analysis identifies foundational contributions centered on algorithmic fairness, accountability, explainability, and governance, while historiographic patterns reveal major developmental pathways in credit scoring, financial services, accounting and auditing, and generative AI. Conceptual and thematic analyses further show that the literature is organized into six interconnected clusters covering AI ethics and governance, algorithmic fairness, explainable AI, fraud detection, trustworthy AI, and human-in-the-loop financial decision-making. The study contributes a structured map of this emerging field and shows that AI in finance is increasingly understood not merely as a technical innovation, but as a socio-technical governance challenge requiring responsible design, institutional accountability, and sustained human oversight. © The Author(s) 2026.}, keywords = {Algorithmic decision-making; Artificial intelligence; Bibliometric review; Ethical finances; Explainable AI}, } @article{chenjpangmqianwzhangwxuchep202621, title = {Governing the rise of AI in healthcare: A comparative governance document and implementation implications across five jurisdictions}, author = {Chen J. and Pang M. and Qian W. and Zhang W. and Xu C. and He P.}, year = {2026}, journal = {Social Science and Medicine}, volume = {400}, pages = {119270}, doi = {10.1016/j.socscimed.2026.119270}, abstract = {Background As AI increasingly reshapes healthcare delivery and innovation, global health systems face urgent imperatives to govern its development, deployment. Objective Drawing on 43 regulatory and strategic documents in five jurisdictions, the study systematically assessed convergences and divergences in national AI strategies, risk classification schemes, software-as-a-medical-device (SaMD) frameworks, ethical oversight, and cross-border data policies. Methods Policy and regulatory documents on AI in health (2017–2025) were collected from health authorities, regulatory agencies, and multilateral platforms. A structured 20-item framework covering six domains was applied to enable systematic cross-jurisdictional comparison of strategic vision, regulatory rigor, ethics, implementation support, global alignment, and medical device governance. Results Increasing convergence around core principles such as transparency, accountability, and human-centric AI, particularly through the influence of international norms. However, significant divergence remained in regulatory maturity, enforcement mechanisms, reimbursement pathways and implementation capabilities. While jurisdictions like Singapore and China exhibited centralized, state-led coordination, others like the US reflected pluralistic, agency-driven models. Economic incentives, workforce readiness programs, and post-market surveillance systems also varied widely. Conclusion The study concluded with five recommendations for advancing AI governance: strengthening interagency coordination, harmonizing regulatory frameworks, embedding equity and transparency in data infrastructure, supporting institutional readiness and fostering international collaboration. © 2026 The Authors.}, keywords = {AI; Health policy; Health systems; Healthcare; Artificial Intelligence; China; Delivery of Health Care; Global Health; Health Policy; Humans; Singapore; United States; article; China; economic incentive; global health; health care; health care delivery; health care policy; human; postmarketing surveillance; reimbursement; Singapore; artificial intelligence; health care delivery; health care policy; organization and management; United States}, } @article{maniurathnasamyskamarajk202622, title = {AuditOps: A continuous compliance framework for integrating regulatory requirements in the MLOps lifecycle}, author = {Mani U. and Rathnasamy S. and Kamaraj K.}, year = {2026}, journal = {Future Generation Computer Systems}, volume = {185}, pages = {108649}, doi = {10.1016/j.future.2026.108649}, abstract = {The rapid adoption of machine learning in regulated sectors has revealed a disconnect between conventional MLOps and emerging compliance requirements. We present AuditOps, a framework embedding continuous compliance into the ML lifecycle via four innovations: (1) Regulatory Requirement Embedding Layer; (2) Continuous Explainability Preservation Engine; (3) Automated Audit Trail Generation; and (4) Compliance Drift Detection. We evaluate AuditOps across two domains with different dataset sizes and architectures. In healthcare (Cleveland Heart Disease, 303 instances, Random Forest) under FDA Class II thresholds, AuditOps achieves AUC-ROC of 0.908 (95% CI: 0.879–0.937), CCS = 1.000 (meeting all evaluated FDA Class II thresholds), EPI = 0.904, and ATCM = 1.000. In financial services (UCI Adult Income, 48,842 instances, Multi-Layer Perceptron with three hidden layers of 128-64-32 neurons) under Regulation B fairness constraints, AuditOps achieves AUC-ROC of 0.861 (95% CI: 0.842–0.880), CCS = 0.429 (reflecting realistic class imbalance and disparate impact violations across sex, race, and age flagged by the framework), ATCM = 1.000, and RDDR = 0.750. Statistical validation confirmed significant superiority over random performance across both domains ((Formula presented) ), with AuditOps reducing compliance overhead by 73% compared to MLflow-only baselines. The financial case study (161× larger than healthcare) demonstrates scalability and compatibility with deep learning architectures. We also propose novel compliance metrics (CCS, EPI, ATCM, RDDR) and provide a practical blueprint for integrating regulatory mandates into MLOps workflows, aligning with the forthcoming EU AI Act enforcement in 2026. © 2026 Elsevier B.V. All rights are reserved, including those for text and data mining, AI training, and similar technologies.}, keywords = {AI governance; Audit trails; Continuous monitoring; Deep learning; Explainable AI; MLOps; MLP; Regulatory compliance; Data mining; Embeddings; Health care; Life cycle; AI governance; Audit trails; Class II; Continuous monitoring; Deep learning; Embeddings; Explainable AI; MLOp; MLP; Regulatory requirements; Deep learning; Regulatory compliance}, } @article{lilshirzhangrchenl202623, title = {How can firms leverage responsible artificial intelligence to build competitive advantages? The role of supply chain justice and complexity}, author = {Li L. and Shi R. and Zhang R. and Chen L.}, year = {2026}, journal = {International Journal of Production Economics}, volume = {297}, pages = {110010}, doi = {10.1016/j.ijpe.2026.110010}, abstract = {In highly interconnected supply chains, firms are increasingly adopting artificial intelligence (AI) to improve their forecasting, resource allocation, and responsiveness. However, opaque and bias-prone algorithms can obscure decision-making processes and undermine collaboration between supply chain partners. Responsible AI, in which AI is designed, developed, and deployed in such a way that ensures its ethical conduct, transparency, and alignment with human values, offers a potential remedy, but its relational effects in supply chains remain unclear. Drawing on organizational justice theory, we test whether supply chain justice, encompassing both distributive and procedural justice, mediates the responsible AI–competitive advantage link, with supply chain complexity as a boundary condition. Survey data from 218 Chinese firms show that responsible AI fosters both distributive and procedural justice, which in turn facilitates firms’ competitive advantages. Further, we identify an asymmetric moderating effect: high supply chain complexity weakens the mediating effect of distributive but not procedural justice. Our study advances the research on AI-enabled supply chains by identifying how ethical AI practices can foster firm performance via supply chain justice and when supply chain complexity weakens this effect. Practically, our findings suggest that in complex supply networks, ensuring transparent and consistent decision-making processes provides a more robust mechanism for sustaining competitive advantages compared with focusing only on outcome allocation. Copyright © 2026. Published by Elsevier B.V.}, keywords = {Competitive advantages; Organizational justice theory; Responsible AI; Supply chain complexity; Supply chain justice; Artificial intelligence; Behavioral research; Competitive intelligence; Decision making; Decision theory; Ethical aspects; Supply chains; Competitive advantage; Decision-making process; Organisational; Organizational justice theory; Procedural justice; Resources allocation; Responsible artificial intelligence; Supply chain complexity; Supply chain justice; Supply chain partners; Competition}, } @article{raskarrchopraaalsalamahsfarlowakalraksinghrzhaoygeorgeraroracnouhmshanxcamposspignataronyunltpujarislabriqueabazb202624, title = {Driving global health equity with artificial intelligence: the global initiative on AI for health (GI-AI4H)}, author = {Raskar R. and Chopra A. and AlSalamah S. and Farlow A. and Kalra K. and Singh R. and Zhao Y. and George R. and Arora C. and Nouh M. and Shan X. and Campos S. and Pignataro N. and Yun L.T. and Pujari S. and Labrique A. and Baz B.}, year = {2026}, journal = {npj Health Systems}, volume = {3}, number = {1}, pages = {38}, doi = {10.1038/s44401-026-00089-w}, abstract = {The Global Initiative on Artificial Intelligence for Health (GI-AI4H) is a World Health Organization-led collaboration with the International Telecommunication Union and the World Intellectual Property Organization to support safe, ethical, and equitable Artificial Intelligence (AI) adoption in health. This perspective introduces the RISE framework (Robust capacity building, Inclusive research, Smart infrastructure, and Equitable data practices) as a pragmatic, adaptable approach for advancing responsible AI implementation across diverse health system contexts globally. © The Author(s) 2026.}, } @article{parkyj202625, title = {Algorithmic representation in virtual realities: ethical challenges and regulatory opportunities}, author = {Park Y.J.}, year = {2026}, journal = {Ethics and Information Technology}, volume = {28}, number = {2}, pages = {19}, doi = {10.1007/s10676-026-09895-0}, abstract = {This article conceptualizes the notion of algorithmic representation to illuminate key areas of ethical concern that virtual reality (VR)-based social media prompt to address. We highlight how VR-based social media, as in the ongoing transformation of Facebook and other social media in their integration with artificial intelligence (AI), necessitate reappropriation of regulatory apparatuses of personal data. Doing so, this study contributes to the understanding of metaverse, a VR-based algorithmic platform, as ‘a representational system’, and integrates a virtual reality research perspective with critical algorithm studies to identify highly arbitrary, selective nature of algorithmic representation. We identify the key opportunities of algorithmic governance where the momentum toward VR-based social media must replant its conceptual terrain: (1) cross-platform, (2) cross-time, (3) derivative, (4) marginalized, and (5) concentrated representation. This study’s thesis is that the Silicon Valley industry, as well as its regulators, should critically reflect on unsubstantiated epistemological claim of AI machine-human equivalence. This study argues for the alternative narrative of algorithmic constructedness that is based on participatory representation, rather than accepting the premise of any foreseeable replacement, supplement, and reflection of social actors by an objective machine. © The Author(s), under exclusive licence to Springer Nature B.V. 2026.}, keywords = {AI; AI ethics and regulation; VR representation; VR-based social media; Ethical technology; Metadata; Social networking (online); Virtual addresses; Virtual reality; Algorithm study; Algorithmics; Artificial intelligence ethic and regulation; Cross-platform; Ethical concerns; Facebook; Metaverses; Social media; Virtual reality representation; Virtual reality-based social medium; Artificial intelligence}, } @article{peterdpetermpeterpyacobp202626, title = {Governing intelligent supply chains: Policy directions for AI-driven logistics in the era of green digital transformation}, author = {Peter D. and Peter M. and Peter P. and Yacob P.}, year = {2026}, journal = {Research in Transportation Business and Management}, volume = {68}, pages = {101741}, doi = {10.1016/j.rtbm.2026.101741}, abstract = {This study examines the evolving policy and governance processes that are shaping the adoption of AI-based logistics in Malaysia, within the broader context of the green digital transformation. It measures the correspondence of national strategies, addresses governance and ethical concerns, and provides a contextualized framework of policy to support responsible innovation, aligning with Malaysia's sustainability objectives. A mixed-methods research design was employed, combining Python-based computational topic modeling with NVivo-facilitated thematic analysis of 17 semi-structured interviews. This approach enabled the identification of five policy clusters quantitatively and seven themes qualitatively, which mirror the institutional, ethical, and operational dynamics within the Malaysian logistics ecosystem. The results indicate that Malaysia has a high level of technological ambition and a low level of institutional coherence in establishing ethics and sustainability in AI logistics governance. There are indications of divided responsibility, rhetorical sustainability practice, and the lack of enforceable ethical norms. Additionally, the conceptualisation of the study views this evolution as a paradigmatic shift in digitalisation, focusing on the development of a viable, responsible innovation ecosystem where technology, ethics, and sustainability are intertwined. Lastly, the study suggests the AI-Green Governance Framework, which consists of a centralized AI-Green Governance Council, fiscal incentives, demands for running algorithms transparently, and an AI Sustainability Index (AISI) to assess performance and responsibility in green areas. This study contributes to theoretical and policy discussions by introducing AI governance, ethics, and sustainability as a single policy framework. © 2026 Elsevier Ltd. All rights are reserved, including those for text and data mining, AI training, and similar technologies.}, keywords = {AI-driven logistics; And sustainable supply chain policy; Ethical AI; Green digital transformation; Responsible innovation governance}, } @article{katalshovdshvetsovaoleeskkoltuns202627, title = {Decoupling Intelligence from Governance: A Dynamic Bilateral Architecture for Real-Time Enterprise AI Compliance}, author = {Katalshov D. and Shvetsova O. and Lee S.-K. and Koltun S.}, year = {2026}, journal = {Electronics (Switzerland)}, volume = {15}, number = {10}, pages = {2125}, doi = {10.3390/electronics15102125}, abstract = {The widespread adoption of Generative Artificial Intelligence (GenAI) in regulated enterprises is currently hindered by the “Static Alignment Trap”: the inability of traditional safety methods, such as Reinforcement Learning from Human Feedback (RLHF), to adapt to rapidly shifting compliance landscapes without costly retraining. This paper introduces and evaluates the Agreement Validation Interface (AVI), a modular governance architecture that functions as a deterministic middleware layer. By decoupling governance from the core inference engine, AVI implements Dynamic Bilateral Alignment (DBA), enforcing policy constraints at both the input and output stages through vector-based semantic retrieval. Adopting a Design Science Research (DSR) methodology, we validated the system against the FinanceBench financial benchmark ((Formula presented.) queries, three repeated runs, 450 total observations) and a proprietary Russian-language provocative content dataset developed internally at MWS AI ((Formula presented.) queries; not publicly available). The empirical results demonstrate that the architecture achieves an 83.2% Large Language Model (LLM)-judge compliance rate (95% confidence interval, CI: 79.4–87.1%), statistically significantly exceeding the unfiltered baseline of 63.7% ((Formula presented.) percentage points (pp), (Formula presented.), (Formula presented.)). The vector-based input filter achieves perfect detection performance (Precision (Formula presented.), Recall (Formula presented.), F1 (Formula presented.)). Cross-domain validation on 201 Russian-language provocative queries confirms generalizability (Recall (Formula presented.), LLM compliance among triggered queries (Formula presented.)). The operational Time-to-Compliance for enforcing new rules was reduced from hours (model fine-tuning) to under five seconds (vector indexing). These findings suggest that enterprise AI safety requires an architectural shift from model-centric training to system-centric control, complemented by system-prompt-level anti-inference governance. We conclude that AVI offers a scalable, cost-effective, and statistically validated framework for auditable AI compliance, independent of the underlying model provider. © 2026 by the authors.}, keywords = {AI governance; AI TRiSM; Design Science Research; dynamic alignment; FinTech; Retrieval-Augmented Generation (RAG); Architecture; Benchmarking; Cost effectiveness; Inference engines; Memory architecture; Middleware; Reinforcement learning; Safety engineering; Semantics; AI governance; AI TRiSM; Decouplings; Design-science researches; Dynamic alignment; Enterprise IS; Language model; Real- time; Retrieval-augmented generation; Russian languages; FinTech}, } @article{hameedzalsalehhmalikpvishnoisk202628, title = {How does responsible AI influence healthcare employee AI collaboration and engagement? The role of AI trust and responsible leadership}, author = {Hameed Z. and Alsaleh H. and Malik P. and Vishnoi S.K.}, year = {2026}, journal = {Technological Forecasting and Social Change}, volume = {227}, pages = {124651}, doi = {10.1016/j.techfore.2026.124651}, abstract = {AI technologies have transformed business operations, improving efficiency and effectiveness. Responsible Artificial Intelligence (RAI) prioritises ethics and responsibility in AI development and deployment, creating a more trustworthy and transparent organization. However, there is a paucity of studies regarding the effects of RAI on healthcare employee outcomes, particularly AI engagement. This study examines how RAI influences AI trust, which, in turn, drives AI collaboration and AI engagement of healthcare employees. We also examine the moderating role of responsible leadership on RAI and AI trust. We obtained data from healthcare organisations to test our hypotheses. Drawing on the Stimulus-Organism-Response (S-O-R) framework, our results indicate that AI trust mediates the influence of RAI on employee AI collaboration and AI engagement. Our research highlights the crucial role of trust in AI implementation, suggesting that firms that prioritize RAI practices are more likely to cultivate a motivated and engaged workforce. These conclusions offer significant implications for management practices and future research on RAI and employee outcomes in healthcare. © 2026 Elsevier Inc.}, keywords = {AI collaboration; AI engagement; AI trust; Responsible artificial intelligence; Responsible leadership; Health care; Personnel; AI collaboration; AI engagement; AI Technologies; AI trust; Business operation; Healthcare organizations; Improving efficiency; Management practises; Responsible artificial intelligence; Responsible leaderships; artificial intelligence; health care; leadership; organizational framework; Artificial intelligence}, } @article{lols202629, title = {Artificial intelligence regulation matures: Landscapes of the USA, European Union, and China}, author = {Lo L.S.}, year = {2026}, journal = {IFLA Journal}, volume = {52}, number = {2}, pages = {160-166}, doi = {10.1177/03400352251384915}, abstract = {Between 2023 and July 2025, artificial intelligence (AI) governance in the USA, European Union, and China shifted from programmatic statements to actionable instruments. The USA moved from Executive Order 14110 to three July 2025 executive orders on data-center permitting, export promotion, and procurement neutrality. The European Union completed the AI Act, initiated staged application in 2025, and issued a code of practice for general-purpose AI. China consolidated domestic controls on public-facing generative AI and launched a Global AI Governance Action Plan with United Nations-centered cooperation, standards work, and capacity-building. The UK continued a regulator-led, assurance-first model. This essay compares these trajectories and distils implications for libraries: stronger accountability in procurement and vendor management; lawful, well-described training data; the publication of assessment artifacts; and AI literacy as a core service. The analysis highlights convergence on safety, transparency, and inclusion, alongside divergence in regulatory technique and international posture, which will shape library strategy. © The Author(s) 2025}, keywords = {AI; AI literacy; Artificial intelligence policy; data governance; European Union AI Act; Executive Order 14110; governance; libraries; procurement}, } @article{hussainsqaziawilkv202630, title = {More than just fair: Legitimizing AI through reciprocity, ethical AI characteristics and personal information sharing}, author = {Hussain S. and Qazi A. and Wilk V.}, year = {2026}, journal = {Technovation}, volume = {155}, pages = {103572}, doi = {10.1016/j.technovation.2026.103572}, abstract = {This research examines how perceived reciprocity and perceived ethical AI characteristics – fairness, accountability, and transparency (FAT principles) – shape consumers’ willingness to share personal information and its downstream consequences in AI-mediated consumer-brand interactions. Grounded in social exchange theory, we conceptualize personal information sharing as a behavioral gateway through which ethical exchange cues translate into perceived algorithmic legitimacy and intention to co-create value. Using a two-study design, consisting of a scenario-based experiment (n = 184) and a cross-sectional survey (n = 612), the findings of this research show that reciprocity and FAT principles robustly increase willingness to share personal information. Their effects on perceived algorithmic legitimacy and co-creation intentions operate primarily through information sharing rather than directly. Results further indicate that willingness to share personal information functions as a legitimacy-conferring act, supporting a bottom-up, interactional view of legitimacy formation in AI-mediated exchanges. By distinguishing information sharing-based participation from downstream value co-creation, this study advances a process-based account of engagement and offers actionable insights for designing ethically grounded AI systems that encourage voluntary data sharing and sustained consumer collaboration. © 2026 The Authors}, keywords = {AI; Cross-sectional survey; Ethical AI; Experimental design; Mixed method; Reciprocity; Social exchange theory; Willingness to share personal information; Data Sharing; Design of experiments; Ethical technology; Information analysis; Information dissemination; Statistics; Cross-sectional surveys; Down-stream; Ethical AI; Information sharing; Mixed method; Personal information; Reciprocity; Social exchange theory; Willingness to share; Willingness to share personal information; Artificial intelligence}, } @article{rigotticalvesfernandesdmutpiaafoschvillarongae202631, title = {Transparency in the datafied workplace: Law, workers, and job applicant perspectives}, author = {Rigotti C. and Alves Fernandes D. and Mut Piña A. and Fosch-Villaronga E.}, year = {2026}, journal = {Technology in Society}, volume = {87}, pages = {103387}, doi = {10.1016/j.techsoc.2026.103387}, abstract = {As AI systems become increasingly embedded in the workplace, and the datafication of recruitment and employment grows more pervasive, ensuring the trustworthiness of these technologies from design to deployment has emerged as a pressing concern - particularly within the European Union's evolving regulatory framework. The General Data Protection Regulation (GDPR) and the AI Act play a key role in shaping the governance of AI, including in recruitment and employment. Drawing on a large-scale survey conducted as part of the BIAS project, our study analyses 4317 valid responses from a diverse sample across the European Union, Iceland, Norway, Switzerland, and Turkey. It examines how job applicants and workers perceive and experience AI systems, their awareness of AI use, and their experiences with data processing and transparency mechanisms, paying particular attention to demographic patterns related to gender, age, and education. Our findings reveal that transparency measures, while sometimes present, remain fragmented and procedural. Opt-out mechanisms are also frequently unclear. AI systems in recruitment contexts collect substantially more data, particularly on sensitive topics. In workplace settings, data collection is lower and differences between sensitive and non-sensitive information are less pronounced. Subjective minority identification appears to be the most consistent demographic predictor of higher likelihood of data collection. Building on these findings, the article critically assesses how transparency is, and should be, operationalized under the GDPR and the AI Act. It argues for aligning legal compliance with practices that centre the experiences of those most affected and address structural inequalities in recruitment and employment. © 2026 The Authors.}, keywords = {AI act; Datification; GDPR; Hiring; Transparency; Workplace; Data acquisition; Employment; International law; Population statistics; Sensitive data; AI act; AI systems; Data collection; Datification; European union; General data protection regulations; Hiring; Job applicant; Workers'; Workplace; artificial intelligence; demographic method; recruitment (employment); survey method; workplace; Transparency}, } @article{mccormacklbendechachem202632, title = {The Trustworthy AI Maturity Model (TAIMM): Integrating ethics and regulation across the AI lifecycle}, author = {McCormack L. and Bendechache M.}, year = {2026}, journal = {Journal of Responsible Technology}, volume = {26}, pages = {100156}, doi = {10.1016/j.jrt.2026.100156}, abstract = {This paper introduces the Trustworthy AI Maturity Model (TAIMM), a lifecycle-based evaluation framework designed to support the development of ethically aligned and legally compliant Artificial Intelligence (AI) systems. TAIMM responds to the implementation gap in current AI governance approaches, which often endorse high-level ethical principles but lack tools for operationalising them in practice. Focusing on high-risk systems under the European Union’s AI Act, the model maps the Act’s recitals to the seven principles of Trustworthy AI and integrates them into a structured governance framework grounded in established System Development lifecycle models. Unlike general AI management standards such as ISO 42001, TAIMM provides a maturity-oriented diagnostic tool tailored to the AI lifecycle. The framework includes three stage-specific questionnaires covering design, development, and operation each item explicitly labelled with its corresponding ethical principle. This approach enables both actionable self-assessment and quantitative analysis of ethical coverage, revealing disparities in emphasis and highlighting underrepresented principles. By translating abstract regulatory and ethical expectations into practical, auditable instruments, TAIMM advances responsible AI governance through a scalable, transparent, and context-aware evaluation model. © 2026 Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license. http://creativecommons.org/licenses/by-nc-nd/4.0/}, keywords = {Ethical AI evaluation; EU AI act; ISO 42001; Trustworthy AI governance}, } @article{herrerafcaldernr202633, title = {Opacity as a feature, not a flaw: Role-sensitive explainability, institutional trust, and the LoBOX ethics governance framework for AI}, author = {Herrera F. and Calderón R.}, year = {2026}, journal = {Technology in Society}, volume = {86}, pages = {103302}, doi = {10.1016/j.techsoc.2026.103302}, abstract = {This paper introduces the LoBOX (Lack of Belief: Opacity & eXplainability) ethics governance framework, a governance-centric approach for managing artificial intelligence (AI) opacity when full transparency is infeasible. While transparency-centric approaches treat transparency as the social/ideal goal and therefore opacity as a design flaw, LoBOX suggests opacity is a condition which should be ethically governed through role-sensitive explanation and institutional accountability. The LoBOX framework comprises a three-stage pathway: reduce accidental opacity, bound irreducible opacity, and delegate trust through institutional oversight. Integrating the stakeholder-sensitive explanation described in the RED/BLUE XAI model, which is aligned with emerging legal instruments such as the EU AI Act, LoBOX offers a scalable and context-aware alternative to transparency-centric approaches. LoBOX reframes trust as an outcome of institutional credibility, structured justification, and stakeholder-sensitive accountability, and it is designed to remain aligned with evolving technological contexts and stakeholder expectations while ethically governing opacity. In the end, to ensure responsible AI systems, LoBOX moves from transparency ideals to ethical governance, emphasizing that trustworthiness in AI must be institutionally grounded and contextually justified. Copyright © 2026. Published by Elsevier Ltd.}, keywords = {Accountability; AI governance; Algorithmic opacity; Explainable artificial intelligence (XAI); Institutional trust; Role-sensitive explainability; Artificial intelligence; Ethical technology; Opacity; Accountability; Algorithmic opacity; Algorithmics; Artificial intelligence governance; Condition; Context-Aware; Explainable artificial intelligence (XAI); Institutional trust; Legal instruments; Role-sensitive explainability; accountability; algorithm; artificial intelligence; ethics; governance approach; institutional framework; Transparency}, } @article{senadheerasyigitcanlartdesouzakmossbergerkcheongphcorchadojpaza202634, title = {Is responsible AI a public demand? Trust, risk, and persuasion in local government AI}, author = {Senadheera S. and Yigitcanlar T. and Desouza K. and Mossberger K. and Cheong P.H. and Corchado J. and Paz A.}, year = {2026}, journal = {Information Processing and Management}, volume = {63}, number = {8}, pages = {104932}, doi = {10.1016/j.ipm.2026.104932}, abstract = {Local governments are increasingly integrating AI into public service delivery; however, public acceptance remains uneven and is shaped by multiple factors. This study investigates how trust, persuasion, and perceived risk interact to shape public demand for responsible AI. Using a cross-national survey of 1220 respondents across Australia, Spain, and the United States and analysing the data through structural equation modelling (SEM), the study identifies key psychological drivers of public expectations. The findings reveal that trust consistently reduces perceived AI risk, while persuasion primarily strengthens trust rather than directly influencing risk perceptions. Notable cross-country differences emerged; in Australia, trust plays a dominant role in shaping perceptions of AI risk ((Formula presented) ); in Spain, persuasion ((Formula presented) ) balances for low institutional trust; and in the United States, mediated persuasion accounts for nearly 30% of variance in trust ((Formula presented) ). Across all contexts, perceived risk is positively associated with stronger expectations for responsible AI, explaining 11%–24% of the variance in demand. This study contributes by moving beyond normative policy framing of responsible AI and showing empirically that it is a locally interpreted public demand. This highlights that responsible AI is not a universal construct and calls for context-specific strategies that align with local intuitional trust dynamics and mediated persuasive environments. © 2026 The Author(s). Published by Elsevier Ltd. This is an open access article under the CC BY license. http://creativecommons.org/licenses/by/4.0/}, keywords = {Artificial intelligence; Digital public governance; Local government; Persuasive communication; Public trust; Responsible AI adoption; Risk perception; Government data processing; Public policy; Risk perception; Australia; Digital public governance; Government IS; Local government; Perceived risk; Persuasive communication; Public demands; Public service deliveries; Public trust; Responsible AI adoption; Artificial intelligence}, } @article{laviolafcucarin202635, title = {From promise to concern: Public perceptions of AI in ESG frameworks over time}, author = {Laviola F. and Cucari N.}, year = {2026}, journal = {Technology in Society}, volume = {85}, pages = {103219}, doi = {10.1016/j.techsoc.2026.103219}, abstract = {This study investigates how public sentiment toward Artificial Intelligence (AI) has evolved at the intersection of Environmental, Social, and Governance (ESG) frameworks and the rising field of Corporate Digital Responsibility (CDR) over the past 25 years. Drawing on a dataset of 33,628 news articles published between 2000 and 2025, we conduct a large-scale longitudinal sentiment analysis to identify discursive patterns in the perception of AI's role across the ESG dimensions. Our findings reveal substantial variation across the three pillars. While sentiment toward AI in governance contexts shows a consistently positive trend, associated with increased expectations for transparency, monitoring, and compliance, environmental sentiment exhibits a sharp downturn after 2022, reflecting concerns over the carbon footprint of generative AI technologies. The social dimension displays fluctuating sentiment, influenced by debates on automation, fairness, and ethical accountability. These differentiated trajectories suggest that AI legitimacy is a domain-specific and socially negotiated construct, rather than a uniform outcome of technological advancement. Public discourse, as captured in news media, functions as an anticipatory indicator of emerging regulatory tensions and reputational risks, offering valuable foresight for corporate and institutional decision-makers. This study contributes to the literature on technology and society by highlighting the role of sentiment dynamics in shaping AI governance and sustainability strategies. It provides both theoretical insights into the social construction of technological legitimacy and practical implications for the design of responsive, context-sensitive ESG policies in the age of digital transformation. © 2026 The Authors.}, keywords = {Artificial intelligence; Corporate digital responsibility; Corporate governance; ESG; Natural language processing; Sentiment analysis; Decision making; Environmental technology; Ethical aspects; Industrial management; Information systems; Public risks; Sustainable development; Corporate digital responsibility; Corporate governance; Corporates; Environmental, social, and governance; Language processing; Natural language processing; Natural languages; Public perception; Public sentiments; Sentiment analysis; artificial intelligence; corporate social responsibility; environmental assessment; governance approach; perception; public attitude; Artificial intelligence; Sentiment analysis}, } @article{parkmj202636, title = {Generative AI and responsible supply chains: When sustainability gains undermine social legitimacy}, author = {Park M.J.}, year = {2026}, journal = {Technology in Society}, volume = {85}, pages = {103240}, doi = {10.1016/j.techsoc.2026.103240}, abstract = {Generative artificial intelligence (GenAI) is increasingly adopted in supply chains to advance sustainability goals; however, empirical evidence on its environmental and social consequences remains limited and fragmented. This study examines how GenAI adoption affects responsible supply chain outcomes by integrating Difference-in-Differences (DiD) and fuzzy-set Qualitative Comparative Analysis (fsQCA). Using firm–quarter panel data from 320 South Korean firms covering the period from 2023Q1 to 2025Q2, we estimate the short-run causal effects of GenAI adoption on environmental, operational, and social performance, while also identifying the organizational configurations under which these effects emerge. The DiD results provide causal evidence that GenAI adoption significantly reduces CO2 intensity and strengthens circular supply chain practices in the short term. Mediation analyses further show that these environmental benefits are realized primarily through enhanced circular practices rather than through direct improvements in operational efficiency. In contrast, the average effects on operational efficiency and social responsibility outcomes are statistically insignificant, indicating substantial heterogeneity across firms. The fsQCA analysis uncovers multiple sufficient pathways to sustainability. GenAI adoption combined with high absorptive capacity consistently produces environmental gains, whereas the combination of GenAI, responsible AI governance, and strong managerial commitment is the only configuration that generates joint environmental and social outcomes. Conversely, GenAI adoption in the absence of governance safeguards leads to environmental improvements while undermining social legitimacy, underscoring the conditional and double-edged nature of GenAI in supply chains. By integrating causal inference with configurational analysis, this study advances understanding of when and how GenAI contributes to responsible supply chains. The findings demonstrate that while GenAI can serve as a powerful enabler of circular and low-carbon supply chains, its broader social legitimacy critically depends on complementary organizational capabilities and governance arrangements. © 2026 Elsevier Ltd.}, keywords = {Circular economy; Difference-in-Differences; fsQCA; Generative artificial intelligence (GenAI); Responsible supply chains; South Korea; Artificial intelligence; Circular economy; Economic and social effects; Environmental management; Social aspects; Supply chains; Sustainable development; Circular economy; Difference-in-differences; Differences-in-differences; Environmental consequences; Fuzzy Set Qualitative Comparative Analysis; Generative artificial intelligence; Operational efficiencies; Panel data; Responsible supply chain; Social consequences; artificial intelligence; carbon dioxide; circular economy; comparative study; efficiency measurement; environmental economics; governance approach; panel data; supply chain management; sustainability; Efficiency}, } @article{ibrahimagourarifalqodsie202637, title = {Artificial intelligence between legal governance and professional ethics: A comparative study}, author = {Ibrahim A. and Gourari F. and AlQodsi E.}, year = {2026}, journal = {Social Sciences and Humanities Open}, volume = {13}, pages = {102841}, doi = {10.1016/j.ssaho.2026.102841}, abstract = {This study explores the intersection between artificial intelligence (AI) governance and professional ethics through a comparative legal analysis of Egypt and the United Arab Emirates (UAE). With the growing reliance on AI in professional practice, its influence on ethical standards particularly regarding confidentiality, professional integrity, and the independence of decision-making has become a matter of considerable concern. The study identifies key challenges arising from AI integration, including data security risks, algorithmic bias, and the potential erosion of professionals’ independent judgment. Within a comparative legal framework, the research examines the regulatory measures currently in force in Egypt and the UAE, assessing their effectiveness in safeguarding ethical conduct in professional use of AI. The study concludes by underscoring the need to establish balanced legal frameworks that promote technological innovation while ensuring ethical accountability. It also offers recommendations for reinforcing AI governance, enhancing legal oversight, and preserving adherence to professional values in AI-dependent work environments. Copyright © 2026. Published by Elsevier Ltd.}, keywords = {Artificial intelligence (AI); Confidentiality; Impartiality- civil liability- digital fabrication; Integrity; Legal governance; Professional ethics}, } @article{shimizuhpuzioamamakkgunkeldj202638, title = {Who belongs in the moral circle? AI systems and the limits of inclusion}, author = {Shimizu H. and Puzio A. and Mamak K. and Gunkel D.J.}, year = {2026}, journal = {Asian Journal of Philosophy}, volume = {5}, number = {2}, pages = {57}, doi = {10.1007/s44204-026-00416-w}, abstract = {This paper offers a relational response to Jeff Sebo’s The Moral Circle and its moral-expansionist framework under conditions of uncertainty. While Sebo argues that we should consider not only which beings matter but which beings might matter, and extends this precautionary logic to animals, AI systems, and other marginal cases, we argue that this framework leaves underexamined the standpoint from which moral-circle expansion is undertaken. A relational approach does not reject the importance of avoiding moral exclusion. Rather, it reframes the question by asking what it means to “expand” the moral circle in the first place, how moral circles are historically and socially constituted, and who has the authority to draw and redraw their boundaries. On this view, moral significance does not arise solely from intrinsic properties such as sentience, agency, or welfare capacity, but also from the relations, institutions, practices, and power structures through which beings become ethically salient. We therefore ask not simply whether AI belongs within the moral circle, but how the very framing of inclusion is shaped when AI systems are in view. We argue that Sebo’s critique of human exceptionalism remains partially constrained by a human standpoint that continues to administer inclusion. The paper concludes that the ethical challenge posed by AI is not only to determine whether artificial systems belong within the moral circle, but also to examine how moral relations are formed, mediated, and governed in increasingly complex sociotechnical worlds. © The Author(s), under exclusive licence to Springer Nature B.V. 2026.}, keywords = {AI ethics; Artificial intelligence; Human exceptionalism; Moral circle; Relational approach}, } @article{mathessferberddreyertbormkjmodersohnlwillemtdirvenrvibertjkreutzfeldtsperezlopezrprelajastrandfbairdrdboekermkatherjntschochoheimlammertj202639, title = {Collaborative framework on responsible AI in LLM-driven CDSS for precision oncology leveraging real-world patient data}, author = {Mathes S. and Ferber D. and Dreyer T. and Borm K.J. and Modersohn L. and Willem T. and Dirven R. and Vibert J. and Kreutzfeldt S. and Perez-Lopez R. and Prelaj A. and Strand F. and Baird R.D. and Boeker M. and Kather J.N. and Tschochohei M. and Lammert J.}, year = {2026}, journal = {npj Precision Oncology}, volume = {10}, number = {1}, pages = {15}, doi = {10.1038/s41698-025-01180-5}, abstract = {Precision oncology leverages real-world data, essential for identifying biomarkers and therapies. Large language models (LLMs) can aid at structuring unstructured data, overcoming current bottlenecks in precision oncology. We propose a framework for responsible LLM integration into precision oncology, co-developed by multidisciplinary experts and supported by Cancer Core Europe. Five thematic dimensions and ten principles for practice are outlined and illustrated through application to uterine carcinosarcoma in a thought experiment. © The Author(s) 2025.}, keywords = {Article; awareness; clinical decision support system; cognitive bias; computer security; conceptual framework; consensus; consent; ethics; Europe; fairness; human; infrastructure; large language model; medical education; multidisciplinary team; patient coding; personalized cancer therapy; privacy; protocol compliance; regulatory compliance; responsible artificial intelligence; risk benefit analysis; thematic analysis; uterine carcinosarcoma}, } @article{wrightcs202640, title = {Training on the tightrope: AI copyright and data privacy as colliding regulatory regimes}, author = {Wright C.S.}, year = {2026}, journal = {Computer Law and Security Review}, volume = {61}, pages = {106343}, doi = {10.1016/j.clsr.2026.106343}, abstract = {Artificial-intelligence systems are trained on datasets that simultaneously implicate two bodies of law operating on incompatible assumptions. Copyright doctrine treats training data as expressive works subject to use-based analysis, evaluating ingestion through transformative purpose, market substitution, and the nature of the copied work. Data-privacy law treats the same data as personal information subject to identity-based rights, demanding lawful bases for processing, honouring erasure requests, and constraining automated decision-making. This article identifies and maps five structural contradictions arising when the regimes collide in the AI training context — the Retention Paradox, the Unlearning Impossibility, the Consent–Licence Gap, the Jurisdictional Fracture, and the Remedial Mismatch — derives them through comparative doctrinal analysis, and tests them against hybrid and risk-based regulatory proposals already advanced in the AI governance literature. Drawing on a comparative analysis of the United States and European Union (with a methodological note on why those jurisdictions, and how the framework extends to others), it argues that doctrinal adjustment within either regime cannot resolve the contradictions, because they arise from different ontologies of what “data” is. The article proposes an interstitial regulatory framework — a “law of training data” — built on a unified regulatory category, a tiered compliance model, a model-audit regime drawing on incident-reporting and federated-governance designs, statutory safe harbours, and a dedicated coordination mechanism. The article specifies operational design (statutory text, regulatory placement, enforcement triggers) for each pillar and engages the technical machine-unlearning literature to distinguish what is technically infeasible from what is economically impractical. © 2026 The Author(s)}, keywords = {AI Act; Artificial intelligence; Copyright; Data protection; GDPR; Machine unlearning; Training data; Compliant mechanisms; Data privacy; Decision making; Regulatory compliance; Risk analysis; AI act; Artificial intelligence systems; GDPR; Identity-based; Machine unlearning; Market substitution; Personal information; Privacy law; Regulatory regime; Training data; Artificial intelligence; Copyrights}, } @article{wisedpanichnwittayakoms202641, title = {A conceptual model of employees’ behavioral intention to use generative artificial intelligence technology in mid-sized organizations}, author = {Wisedpanich N. and Wittayakom S.}, year = {2026}, journal = {Multidisciplinary Science Journal}, volume = {8}, number = {8}, pages = {e2026531}, doi = {10.31893/multiscience.2026531}, abstract = {Rapid technological advancements have positioned Generative Artificial Intelligence (GenAI) as a strategic asset for businesses; however, its adoption in resource-constrained environments remains complex. Specifically, in the context of emerging economies where mid-sized firms drive significant employment yet face distinct digital hurdles, understanding these dynamics is crucial. This conceptual research develops a theoretical model explaining employees’ behavioral intention to use GenAI in mid-sized organizations, a sector often overlooked in favor of large corporations or small startups. The study integrates the Technology Acceptance Model (TAM), the Unified Theory of Acceptance and Use of Technology (UTAUT), Adaptive Governance Theory, and the Dynamic Capabilities Framework to capture the multidimensional interplay among organizational, governance, and psychological factors. Unlike traditional models that focus solely on utility, this research posits that organizational conditions and governance conditions act as primary antecedents influencing employees’ perceived risk and trust, which in turn determine behavioral intention toward GenAI use. Eleven research propositions (P1–P11) were formulated to describe both direct and indirect causal relationships, highlighting the mediating roles of psychological safety mechanisms. The study contributes theoretically by extending technology acceptance models beyond cognitive dimensions to include governance and ethical oversight as structural determinants of employee behavior. It also introduces perceived governance as an integrative construct linking organizational readiness to trust and risk perception. Practically, the framework provides actionable guidance for mid-sized organizations to design adaptive governance systems, strengthen transparency, and foster trust-based cultures that encourage responsible GenAI adoption. By highlighting the balance between innovation and accountability, this conceptual model establishes a robust foundation for future empirical validation and policy development aimed at promoting sustainable and ethical AI integration in organizational contexts. Copyright (c) 2026 The Authors. This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License. https://creativecommons.org/licenses/by-nc-nd/4.0/}, keywords = {adaptive governance; organizational readiness; perceived risk; SMEs; technology acceptance; trust}, } @article{zhangszhangtwangx202642, title = {Structural consistency in AI governance: A PMC index assessment with evidence from China’s central-level policies}, author = {Zhang S. and Zhang T. and Wang X.}, year = {2026}, journal = {PLOS ONE}, volume = {21}, number = {6 June}, pages = {e0337024}, doi = {10.1371/journal.pone.0337024}, abstract = {The structural coherence of policy design has become an increasingly important issue in artificial intelligence (AI) governance. This study evaluates the structural consistency of China’s central-level AI policies issued between 2016 and 2025 (n = 54). It combines text mining to identify high-frequency policy terms and semantic co-occurrence patterns with a Policy Modeling Consistency (PMC) index framework comprising nine primary and forty-three secondary indicators. Five representative policies are then selected for detailed quantitative evaluation and visual comparison. The results show that China’s AI policy system is generally well structured, but still exhibits notable weaknesses in temporal planning, intergovernmental coordination, and incentive design. In particular, long-term policy supply remains limited, vertical coordination mechanisms are insufficiently institutionalized, and policy instruments are unevenly configured across key support dimensions. These findings suggest that future policy improvement should focus on strengthening medium- and long-term planning, enhancing coordination across governance levels, and improving the integrated design of policy instruments. Methodologically, the study demonstrates a reproducible analytical framework linking text analysis, indicator construction, quantitative evaluation, and visualization. It contributes to the literature by moving from thematic description toward structural assessment in the study of AI governance. © 2026 Zhang et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.}, keywords = {Artificial Intelligence; China; Data Mining; Humans; Policy Making; artificial intelligence; China; data mining; human; management}, } @article{dotanrgershonithadariluriag202643, title = {Decoupling in AI ethics: Learning how to walk the talk}, author = {Dotan R. and Gershoni T. and Hadar I. and Luria G.}, year = {2026}, journal = {Empirical Software Engineering}, volume = {31}, number = {5}, pages = {131}, doi = {10.1007/s10664-026-10861-z}, abstract = {In recent years, AI ethics declarations, commitments, and frameworks for AI systems development have proliferated. Yet, implementation remains persistently low. This phenomenon, often termed “AI ethics washing,” has been widely criticized but lacks empirical investigation. Our paper examines these gaps between declarations and operations in AI ethics through the organizational psychology concept of “decoupling”—the disconnect between what organizations say and what they do. Using data collected through in-depth interviews with 32 practitioners across diverse companies, from early-stage startups to large corporations, we present a systematic analysis of decoupling between declarations and operations in AI ethics, producing the first analysis of decoupling not only in AI ethics but in any technology development field. Our findings identify and characterize (i) common types of AI ethics declarations, such as policies and internal communications, (ii) common types of AI ethics operations, such as reviews and testing, (iii) common rationales behind companies’ approaches to AI ethics, and (iv) distinct decoupling profiles, i.e., common ways in which AI ethics declarations come apart from operations. Our discussion includes recommendations for increasing AI ethics adoption tailored to each profile. These recommendations differ from traditional AI ethics frameworks. While traditional frameworks prescribe ideal practices based on regulatory or industry expectations, this paper offers recommendations grounded in an empirical analysis of how AI ethics efforts succeed or fail in practice. Our decoupling-informed perspective fundamentally reshapes how practitioners and scholars can approach the challenge of AI ethics implementation. © The Author(s) 2026.}, keywords = {AI Ethics; AI Ethics Washing; AI Governance; Climate; Decoupling; Leadership; AI ethic; AI ethic washing; AI governance; AI systems; Climate; Decouplings; Empirical investigation; Leadership; Organizational psychology; System development; Artificial intelligence}, } @article{hassanamcoertjhclemensmwhassaneinahwaljeejfnelsonjamehrarabjselberjc202644, title = {Governance Framework for Safe and Ethical Implementation of Artificial Intelligence in Surgery: A Modified Delphi Consensus}, author = {Hassan A.M. and Coert J.H. and Clemens M.W. and Hassanein A.H. and Waljee J.F. and Nelson J.A. and Mehrara B.J. and Selber J.C.}, year = {2026}, journal = {Journal of the American College of Surgeons}, volume = {243}, number = {1}, pages = {237-248}, doi = {10.1097/XCS.0000000000001834}, abstract = {BACKGROUND: Artificial intelligence (AI)-enabled clinical decision support systems (CDSS) demonstrate performance comparable or superior to human experts in certain tasks. However, their integration into surgical practice faces a significant implementation gap, alongside ethical, privacy, and legal concerns. Clear governance frameworks are needed to guide their responsible adoption in surgery, to prevent inconsistent application, care quality variation, and exacerbation of algorithmic bias. Here, we establish a systematic, evidence-based, and consensus-driven framework to guide the ethical, effective, and sustainable adoption of AI-enabled CDSS in surgery. STUDY DESIGN: A systematic literature review was conducted of PubMed, Cochrane Library, Medline, and Embase databases until 2024 to identify key governance themes. The themes informed the generation of candidate items, which were then refined through a multiround expert panel consensus process using a modified Delphi approach to produce the final framework. RESULTS: Thematic analysis of 80 full-text articles meeting inclusion criteria identified 4 overarching themes for AI governance: (1) technical prerequisites and model design; (2) clinical implementation and human factors; (3) ethics, safety, and trustworthiness; and (4) bias, fairness, and equity. Panel consensus evaluation resulted in the development of a 19-item framework. CONCLUSIONS: The consensus-driven framework presented here provides foundational guidance essential for navigating the complexities of implementing AI-enabled CDSS safely and ethically in surgery. Addressing the considerations outlined across these 4 core themes can facilitate the responsible adoption of AI, accelerating the transition toward an advanced, data-driven surgical practice while mitigating potential risks. Copyright © 2026 by the American College of Surgeons. Published by Wolters Kluwer Health, Inc. All rights reserved.}, keywords = {Artificial Intelligence; Decision Support Systems, Clinical; Delphi Technique; Humans; artificial intelligence; clinical decision support system; Delphi study; ethics; human}, } @article{adoborh202645, title = {Navigating the AI adoption trap: A framework for organizational practice}, author = {Adobor H.}, year = {2026}, journal = {Organizational Dynamics}, volume = {55}, number = {2}, pages = {101232}, doi = {10.1016/j.orgdyn.2026.101232}, abstract = {Organizations are investing heavily in artificial intelligence (AI), yet many struggle to translate adoption into sustained value. The core challenge is often not technological capability but the AI solutionism trap: a managerial tendency to frame complex organizational problems as primarily solvable through algorithms and data. Drawing on research in managerial decision-making, socio-technical systems, and AI governance, this article explains how solutionism emerges in everyday practice, why it persists, and how it undermines performance, accountability, and trust. Many AI failures are shown to be predictable outcomes of managerial dynamics, including premature problem closure, escalation of commitment, and diffusion of responsibility as systems scale. To mitigate these risks, the paper introduces a practical managerial playbook centered on governance-in-use: concrete practices that slow adoption, reopen problem framing before deployment, make evaluation transparent, and legitimize decisions to pause, modify, or withdraw AI systems. The central implication is clear: sustainable value from AI depends less on speed of adoption than on disciplined judgment under uncertainty. © 2026 The Authors}, keywords = {AI solutionism; Artificial intelligence; Decision-making under uncertainty; Governance-in-use; Managerial judgment; Responsible AI}, } @article{sungkarungsripkiattisins202646, title = {How trustworthy AI fosters adoption and user advocacy in financial advisory services}, author = {Sungkarungsri P. and Kiattisin S.}, year = {2026}, journal = {Discover Artificial Intelligence}, volume = {6}, number = {1}, pages = {455}, doi = {10.1007/s44163-026-01451-5}, abstract = {In many countries, and particularly in developing economies, the availability of professional financial advisors remains far below what is needed. This shortage leaves large numbers of people without structured support for saving, investment, or retirement planning. Those who do seek out human advisors often report strong trust in their expertise, but they represent only a small portion of society. Recent advances in artificial intelligence have opened a different pathway. AI Financial Advisors are not positioned to replace professionals outright but to act as complementary tools that extend affordable and timely guidance to wider groups. Their acceptance, however, depends on whether people regard such systems as ethical and therefore trustworthy. To investigate this question, the study develops a model that brings together the Stimulus–Organism–Behavior–Consequence (SOBC) framework and the Unified Theory of Acceptance and Use of Technology (UTAUT). SOBC is used to trace the process of decision-making: perceptions of ethical design—such as equity and sustainability, privacy and safety, transparency and accountability, shape trust in AI, which then lead to intentions and advocacy. UTAUT complements this by highlighting the role of familiar technology drivers, including effort expectancy, performance expectancy, social influence, and the extent to which users perceive anthropomorphism and personalization in AI systems. Survey data were gathered from 420 employed adults in Thailand who actively use mobile banking applications that incorporate AI-enabled financial advisory features. Results from structural equation modeling confirm that trust in AI acts as the central mechanism. It strongly predicts intention to use and directly influences both word-of-mouth advocacy and customer loyalty. Ethical dimensions that emphasize equity and sustainability, transparency and accountability, privacy and safety strengthen trust. From the technology side, performance expectancy, social influence, and anthropomorphism all had significant effects, while effort expectancy and personalization did not. Overall, the findings show that AI Financial Advisory Services should be understood not only as technological artifacts but as strategic mechanisms for expanding access to financial planning in ways that support inclusion and sustainability. The proposed model connects business strategy with sustainable development and AI governance, offering direction for both policymakers and practitioners in designing advisory systems that are effective in practice and trusted by society. © The Author(s) 2026.}, keywords = {AI ethics; Financial advisory; Trustworthy AI; User adoption; User advocacy; Behavioral research; Economics; Ethical technology; Investments; Social sciences computing; Sustainable development; Advisory services; AI ethic; Financial advisers; Financial advisory; Organism's behavior; Performance; The unified theory of acceptance and use of technology(UTAUT); Trustworthy AI; User adoptions; User advocacy; Decision making}, } @article{mccarthym202647, title = {Responsible Decision Making for AI in Healthcare: Exploring the Role of Ethics Consultants at the Intersection of Ethics and AI}, author = {McCarthy M.}, year = {2026}, journal = {HEC Forum}, volume = {38}, number = {2}, pages = {207-221}, doi = {10.1007/s10730-025-09567-4}, abstract = {Healthcare Ethics Consultants (HECs) can serve as a resource for facilitating values-based conversations for responsible integration of AI technologies that benefit patients, providers, and healthcare organizations. First, the paper describes the different types of AI and its uses in healthcare. Second, it considers the role of HECs and why facilitating conversations around the responsible use and implementation of AI better frames the ethical content for AI healthcare. Third, the paper explores the ethical knowledge necessary to think through responsible use of AI. Finally, it identifies how ethics can be embedded into the process of adopting AI in healthcare from its development to its implementation in the organization and in the need for continual research on its use. The skills necessary for HECs can be utilized in a way to better improve values-based decision-making and evaluation for AI in healthcare. © This is a U.S. Government work and not under copyright protection in the US; foreign copyright protection may apply 2025.}, keywords = {AI ethics; Artificial intelligence; Clinical ethics; Healthcare delivery; Healthcare ethics consultant; Organizational ethics; Artificial Intelligence; Consultants; Decision Making; Delivery of Health Care; Ethicists; Humans; adverse drug reaction; article; artificial intelligence; clinician; controlled study; decision making; ethics; government; health care; health care delivery; health care organization; human; medical ethics; United States; consultation; ethicist; ethics}, } @article{sumritdapiwatvaravongs202648, title = {Prioritizing multi-stakeholder solutions for autonomous delivery robot implementation in urban last-mile logistics: An integrated socio-technical systems and institutional theory framework}, author = {Sumrit D. and Apiwatvaravong S.}, year = {2026}, journal = {Transportation Engineering}, volume = {24}, pages = {100446}, doi = {10.1016/j.treng.2026.100446}, abstract = {Autonomous delivery robots (ADRs) are widely promoted as a solution to improve efficiency in urban last-mile logistics, yet their large-scale adoption remains limited. This is mainly due to social, regulatory, and infrastructure-related challenges rather than technological constraints. This study develops a structured decision-making framework to analyze ADR implementation barriers and to identify effective multi-stakeholder solutions while accounting for uncertainty and interdependencies in expert judgments. The results show that ADR barriers are highly interconnected. Institutional arrangements and infrastructure conditions act as underlying drivers that influence many other challenges, whereas public acceptance, cybersecurity concerns, and human–robot interaction issues largely emerge as outcomes of weak governance and limited system readiness. Safety compliance and risk liability are identified as the most influential barriers, followed by cybersecurity vulnerabilities and cultural or behavioral constraints. These findings indicate that progress in ADR adoption depends more on clear responsibilities, trust, and social readiness than on further technological development alone. The solution analysis shows that governance- and infrastructure-oriented actions are more effective than isolated technical measures. Ethical guidelines, clear legal frameworks, and stronger industry collaboration are identified as the most effective strategies, highlighting the need for coordinated action among regulators, technology developers, and logistics service providers. Theoretically, the study reinforces the view that ADRs should be understood as part of a broader urban system rather than as standalone technologies. Practically, the findings provide clear guidance for stakeholders seeking to enable responsible and scalable ADR deployment in urban last-mile logistics. © 2026 The Author(s)}, keywords = {Autonomous delivery robots; Institutional theory; Multi-criteria decision making; Socio-technical systems; Urban last-mile logistics; Cybersecurity; Decision making; Laws and legislation; Logistics; Robots; Autonomous delivery robot; Institutional theory; Last mile; Multi criteria decision-making; Multi-stakeholder; Multicriteria decision-making; Multicriterion decision makings; Robot implementation; Sociotechnical systems; Urban last-mile logistic; Decision theory}, } @article{spitaleggermanif202649, title = {The making of digital ghosts: designing ethical AI afterlives}, author = {Spitale G. and Germani F.}, year = {2026}, journal = {Ethics and Information Technology}, volume = {28}, number = {3}, pages = {34}, doi = {10.1007/s10676-026-09910-4}, abstract = {The rapid proliferation of AI-mediated digital afterlife technologies, from chatbots trained on personal data to voice clones and posthumous avatars, has generated a substantial body of ethical literature identifying the moral risks of posthumous simulation. Yet this growing consensus has not been matched by frameworks capable of translating ethical principles into operational design constraints. This paper addresses that gap from the perspective of ethical design and governance. We introduce a nine-dimensional taxonomy of digital afterlife technologies, mapping the features that carry independent moral weight in any such system: timing, consent, data sources, interaction modality, fidelity and disclosure, purpose, audience and access, governance and ownership, and autonomy and behavioral agency. Building on this taxonomy, we derive a two-tier structure of design constraints: three threshold constraints — consent, fidelity/disclosure, and purpose — that function as near-absolute conditions of permissibility (Tier 1), and six contextual dimensions that modulate ethical risk profiles without individually determining permissibility (Tier 2). A system that fails any single Tier 1 constraint is impermissible regardless of its Tier 2 configuration. We argue that the ethical constraints derived from the taxonomy are not merely evaluative, but auditable, regulatable, and actionable by designers, governance actors, and legislators. By locating normative assessment at the level of design configuration rather than stated intent alone, the framework offers a concrete bridge between existing ethical consensus and the governance of AI-mediated digital afterlives. © The Author(s) 2026.}, keywords = {AI ethics; Deadbots; Death and technology; Design ethics; Digital ghosts; Postmortem data; Artificial intelligence; Bridges; Ethical technology; AI ethic; Chatbots; Deadbot; Death and technology; Design constraints; Design ethics; Digital ghost; Ethical principles; Moral risks; Postmortem data; Taxonomies}, } @article{sebastiang202650, title = {Digital shadow AI risk theory (DART): A framework for managing data disclosure and privacy risks of AI tools at work}, author = {Sebastian G.}, year = {2026}, journal = {Technological Forecasting and Social Change}, volume = {229}, pages = {124697}, doi = {10.1016/j.techfore.2026.124697}, abstract = {The swift adoption of generative and agentic AI tools in workplace settings has introduced new organizational risks related to data disclosure, privacy, and governance. This study introduces the Digital Shadow AI Risk Framework (DART), which identifies and explains the behavioral and organizational risks arising from the informal and often unregulated use of AI tools by employees. DART comprises six interrelated risk dimensions: unintentional disclosure risk, the trust–dependence paradox, data sovereignty conflict, knowledge dilution, the ethical black box problem, and organizational feedback loops. The framework is evaluated using a three-wave survey research combining hypothesis testing and covariance-based structural equation modeling (CB-SEM) across three cross-industry surveys of professionals (Survey-1: N = 374; Survey-2: N = 179; Survey-3: N = 220). Survey-3 introduced multi-item latent measures enabling direct tests of H4, H6, and the full H8 mediation chain (opacity → trust → comfort → disclosure). Results support six of eight hypotheses. Knowledge dilution is confirmed through replication, while data sovereignty conflict consistently operates as a boundary condition rather than a direct predictor. The findings reveal persistent gaps in employee awareness, training, and organizational controls surrounding AI use. DART's contribution lies in distinguishing Shadow AI from traditional Shadow IT by showing how everyday, efficiency-driven AI use embeds risk into routine knowledge work. By externalizing organizational knowledge into adaptive AI systems, Shadow AI introduces risks that extend beyond technical non-compliance to cognitive dependence and governance erosion. The framework informs future research and supports the development of organizational policies and controls for responsible AI use. © 2026 The Author}, keywords = {Compliance; Data privacy; Generative AI; Organizational risk; Shadow AI risk; Knowledge management; Personnel; Risk management; Statistical tests; Compliance; Disclosure risk; Generative AI; Organisational; Organizational controls; Organizational risk; Privacy risks; Risk frameworks; Risk theory; Shadow AI risk; Data privacy}, } @article{leonm202651, title = {Lifecycle-Based Governance to Build Reliable Ethical AI Systems}, author = {Leon M.}, year = {2026}, journal = {Systems Research and Behavioral Science}, volume = {43}, number = {3}, pages = {1116-1132}, doi = {10.1002/sres.70014}, abstract = {Artificial intelligence (AI) systems represent a paradigm shift in technological capabilities, offering transformative potential across industries while introducing novel governance and implementation challenges. This paper presents a comprehensive framework for understanding AI systems through three critical dimensions: trustworthiness characteristics, lifecycle management, and stakeholder ecosystem. We systematically analyze the technical and operational requirements for robust, reliable, and ethical AI deployment, drawing upon established industry practices while addressing contemporary challenges. The framework emphasizes the dynamic nature of AI systems compared with traditional software, particularly in their data dependencies, continuous learning requirements, and probabilistic outputs. For organizational leaders, we provide actionable insights into risk mitigation, compliance strategies, and governance structures necessary for responsible AI adoption. The paper concludes with strategic recommendations for aligning AI initiatives with business objectives while maintaining ethical standards and regulatory compliance. To enhance practical relevance, the analysis is supplemented with brief case vignettes from manufacturing, finance, healthcare, and public administration, which illustrate how the framework reveals hidden risks and guides effective interventions. The conceptual model and real-world examples offer an integrated roadmap for researchers and practitioners. © 2026 The Author(s). Systems Research and Behavioral Science published by International Federation for Systems Research and John Wiley & Sons Ltd.}, keywords = {accountability; AI governance; ethical AI; lifecycle management; regulation; stakeholder ecosystem; trustworthy AI; Behavioral research; Ethical technology; Life cycle; Public administration; Public risks; Regulatory compliance; Accountability; Artificial intelligence governance; Artificial intelligence systems; Behavioral science; Ethical artificial intelligence; Lifecycle management; Regulation; Stakeholder ecosystem; Systems research; Trustworthy artificial intelligence; Artificial intelligence}, } @article{alshehriksayedmouchawehm202652, title = {Generative AI for the power sector: Methodology, safety, trustworthiness, and policy}, author = {Alshehri K. and Sayed-Mouchaweh M.}, year = {2026}, journal = {Next Energy}, volume = {12}, pages = {100636}, doi = {10.1016/j.nxener.2026.100636}, abstract = {The integration of emerging energy technologies such as renewables, hydrogen fuel cells, battery storage, and increased grid interconnection into power grids introduces significant uncertainty in generation forecasting, grid stability, and demand-supply balancing. Thus, power grids are undergoing a monumental transformation to enable the transition toward sustainable, low-carbon energy systems. They are evolving into digitalized smart grids where cyber, physical, and social domains converge through intelligent components, advanced metering infrastructure, and wireless connectivity. In parallel, recent breakthroughs in artificial intelligence (AI), particularly Generative AI (GenAI), offer transformative technological capabilities useful for addressing these emerging challenges. While existing literature extensively explores individual applications of AI within the power sector, comprehensive assessments of GenAI's multidimensional impacts encompassing operational value, safety, trustworthiness, and regulatory frameworks remain limited. To bridge this gap, we introduce a 3-dimensional analytical framework focused explicitly on operational value creation, safety, and trustworthiness considerations, and tailored policy and regulatory implications. We propose a structured, replicable methodology for constructing GenAI use-cases, demonstrating its effectiveness using predictive maintenance as a representative example, with performance gains exceeding 31% compared to traditional AI methods. Further, we critically assess GenAI safety risks, establish a clear taxonomy with quantifiable metrics for trustworthy AI deployment, and derive targeted policy and regulatory recommendations based on emerging global practices relevant to the power sector. © 2026 The Author(s)}, keywords = {AI governance; AI policy; AI safety; Generative artificial intelligence (GenAI); Power system operations; Predictive maintenance; Smart grid; Trustworthy AI}, } @article{figtalamancag202653, title = {The value of vulnerability for trustworthy AI}, author = {Figà-Talamanca G.}, year = {2026}, journal = {AI and Society}, volume = {41}, number = {5}, pages = {4987-5008}, doi = {10.1007/s00146-026-02892-3}, abstract = {The concept of trustworthy AI plays a significant role in many policy recommendations and ethics guidelines, as it is implied to be a good compromise between the desire for technological innovation and the duty to respect citizens’ rights. However, the concept has often been criticized for its unclarity, its dubious ethical value, and has even been seen as a symptom of ethics washing. I aim to repurpose the value of trustworthy AI for this kind of technology, its development, deployment, and regulation by clarifying the relationship with vulnerability within social arrangements. I argue that vulnerability is one of the key reasons why we enter social arrangements to begin with, and that trust not only works as a facilitator of those social arrangements but requires the recognition and willingness to address vulnerability. Rethinking the trustworthiness of AI and its surrounding practices as a matter of aptly recognizing and addressing vulnerability leads to reconsidering not just the conceptual grounding of trustworthy AI, but also the responsibilities of sovereign actors and the priorities of AI governance. I conclude by exploring the practical implications of such a renewed conception of Trustworthy AI. © The Author(s) 2026.}, keywords = {Governance and Artificial Intelligence; Trust; Trustworthy AI; Vulnerability; Ethical technology; Ethical values; Governance and artificial intelligence; Policy recommendations; Technological innovation; Trust; Trustworthy AI; Vulnerability; Artificial intelligence}, } @article{ahmedzkakarlajhhhazraagurusamym202654, title = {Quantum machine learning for industry 5.0: Fundamental, applications and research challenges}, author = {Ahmed Z. and Kakarla J.H.H. and Hazra A. and Gurusamy M.}, year = {2026}, journal = {Computer Science Review}, volume = {61}, pages = {100976}, doi = {10.1016/j.cosrev.2026.100976}, abstract = {Industry 5.0 is a paradigm shift where human-centric innovation, sustainability, and ethics are the foundation of technology development. This study examines the ability of Quantum Machine Learning (QML) and Internet of Things (IoT) integration to drive the Industry 5.0 revolution through energy efficiency, openness, and secure data management. The innovations comprise green quantum computing for energy savings, quantum explainability for enhancing ethical decision-making, and quantum blockchain for building trust in industrial networks. A five-layer Quantum IoT (QIoT) architecture is proposed, incorporating adaptive feedback and ethical governance, to align with the principles of Industry 5.0. Practical applications like optimizing sustainable supply chains, disaster resilience, and enabling Quantum Digital Twins (QDTs) for smart manufacturing are described. Scaling up nascent quantum technologies while ensuring ethical integrity is also explored in this article, and a roadmap for developing sustainable, human-centric industrial systems is presented. Furthermore, this article addresses the ethical aspects of QML in Industry 5.0, namely, guaranteeing fairness, transparency, and responsible AI use. These challenges need to be surmounted in order to align QML innovations with Industry 5.0’s people-first values. © 2026 Elsevier Inc.}, keywords = {Green quantum computing; Industry 5.0; QIoT architecture; QML; Quantum explainability; Sustainable IoT; Ethical technology; Green computing; Green manufacturing; Industry 4.0; Internet of things; Network security; Quantum computers; Quantum cryptography; Smart manufacturing; Sustainable development; Green quantum computing; Human-centric; Industry 5.0; Machine-learning; Quantum Computing; Quantum explainability; Quantum IoT architecture; Quantum machine learning; Quantum machines; Sustainable internet of thing; Energy efficiency; Information management}, } @article{zhangtqiujxuz202655, title = {Collaborative governance of AIGC disinformation: key factors and strategic dynamics}, author = {Zhang T. and Qiu J. and Xu Z.}, year = {2026}, journal = {Aslib Journal of Information Management}, pages = {1-27}, doi = {10.1108/AJIM-07-2025-0535}, abstract = {Purpose – This study investigates the key factors and evolutionary dynamics of collaborative governance in addressing artificial intelligence-generated content (AIGC) disinformation. Design/methodology/approach – A tripartite evolutionary game model (EGM) is developed to analyze interactions among users, social media, and the government. The model integrates Bounded Rationality Theory (BRT), Deterrence Theory (DT), and Public Goods Theory (PGT). Evolutionary stable strategies are derived using the Jacobian matrix and Lyapunov's first method, supported by numerical simulations and sensitivity analyses. Findings – Social media and the government's initial willingness are vital for accelerating users' transition to compliant behavior through adaptive learning of boundedly rational agents. Due to proximity effects, users are more responsive to social media penalties than government sanctions, revealing an asymmetric deterrence effect. Stronger government regulation reduces social media's incentive for active governance. This creates a free-rider dilemma that requires carefully designed subsidies and penalties to curb opportunistic behavior. Cost-benefit dynamics determine optimal governance – reducing costs or improving effectiveness fosters active governance and lenient regulation, while high costs lead to instability – highlighting the importance of technological innovation for sustainable governance. Originality/value – This study extends EGM and BRT to generative artificial intelligence governance by integrating AI-specific parameters into AIGC disinformation analysis. It refines DT and PGT by revealing asymmetric deterrence effects and formalizing the free-rider dilemma. It proposes a tripartite governance framework that unifies BRT, DT, and PGT and identifies cost-benefit dynamics that challenge the assumption that “more regulation is better.” These findings offer a flexible framework and actionable strategies for collaborative governance of AIGC disinformation. © Emerald Publishing Limited}, keywords = {Artificial intelligence-generated content; Asymmetric deterrence effect; Collaborative governance; Disinformation governance; Evolutionary game model; Free-rider dilemma; Artificial intelligence; Cost benefit analysis; Dynamics; Economic and social effects; Game theory; Government data processing; Jacobian matrices; Public policy; Social networking (online); Artificial intelligence-generated content; Asymmetric deterrence effect; Bounded rationality; Collaborative governances; Disinformation governance; Evolutionary game models; Free-rider dilemma; Free-riders; Public goods; Social media; Sensitivity analysis}, } @article{androutsopouloumcarayannisegaskounisdzotasn202656, title = {Towards AI-Enabled Cyber-Physical Infrastructures—Challenges, Opportunities, and Implications for a Data-Driven eGovernment Theory, Policy, and Practice}, author = {Androutsopoulou M. and Carayannis E.G. and Askounis D. and Zotas N.}, year = {2026}, journal = {Journal of the Knowledge Economy}, volume = {17}, number = {3}, pages = {6170-6207}, doi = {10.1007/s13132-025-02726-5}, abstract = {This article examines the growing intersection of Cyber-Physical Infrastructures (CPI) and Artificial Intelligence (AI) in the context of eGovernment, providing a comprehensive overview of the opportunities, challenges, and policy considerations shaping this convergence. It begins by exploring how AI-driven CPI optimizes data-driven policymaking, public service delivery, and operational efficiency, highlighting use cases such as predictive analytics, resource allocation, and cybersecurity. The discussion then addresses key implementation barriers, including data privacy, algorithmic bias, system integration, and regulatory uncertainties, offering targeted solutions for ethical AI adoption, privacy-preserving methods, and robust cybersecurity frameworks. Emphasizing the potential of emerging technologies such as Quantum AI, federated learning, and blockchain, the article outlines a structured roadmap toward an AI-enabled, resilient, and sustainable CPI. The conclusion underscores the importance of proactive policy measures, interdisciplinary collaboration, and transparent governance in ensuring that AI not only drives innovation but also upholds public trust and democratic values. © The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2025.}, keywords = {AI ethics and regulation; AI-driven governance; Cyber-Physical Infrastructures; EGovernment transformation; Emerging AI technologies; Smart public administration}, } @article{ravendadsalvadojc202657, title = {Janus-faced AI disclosure: How conference call enthusiasm masks workforce setbacks}, author = {Ravenda D. and Salvado J.C.}, year = {2026}, journal = {International Review of Financial Analysis}, volume = {117}, pages = {105255}, doi = {10.1016/j.irfa.2026.105255}, abstract = {Using 71,433 quarterly earnings call transcripts from 4327 U.S.-listed firms (2019–2024), we quantify the intensity and tone of AI-related and employee-centred disclosures with BERTopic and FinBERT, and link them to Refinitiv Social-pillar ESG scores. Panel regressions with firm and industry-year fixed effects, an instrumental variables strategy based on leave-one-out peer disclosure, and staggered difference-in-differences around AI-event calls provide evidence consistent with a plausibly causal interpretation: greater intensity and positivity of AI-related disclosure are followed by lower subsequent workforce and overall social ESG performance. At the same time, firms that trumpet AI most loudly inject a surge of pro-employee rhetoric into those calls, consistent with symbolic “AI-washing.” Taken together, the results suggest a temporal pattern in which firms pair AI announcements with immediate employee-centred rhetoric, while less favourable workforce outcomes emerge later. The findings speak to both legitimacy and stakeholder theories by showing how managers balance innovation signalling with reputational cushioning, using these complementary lenses to explain why firms simultaneously promote AI and emphasize employee-oriented narratives. They also refine the text-as-data paradigm by demonstrating that sentiment-weighted topics from live earnings calls are leading, yet Janus-faced, indicators of social outcomes. Regulators, investors, and rating agencies should therefore triangulate narrative AI enthusiasm with hard labour metrics before inferring ethical AI commitment. © 2026 Elsevier Inc.}, keywords = {AI disclosure; AI-washing; Conference call text analysis; ESG social performance}, } @article{deshazerc202658, title = {The governance gap no one is solving}, author = {DeShazer C.}, year = {2026}, journal = {Journal of the National Medical Association}, volume = {118}, number = {3}, pages = {458-462}, doi = {10.1016/j.jnma.2026.03.009}, abstract = {Artificial intelligence (AI) is rapidly entering clinical practice, yet the governance models needed to ensure its safe, ethical, and equitable use have not kept pace—particularly in community and safety-net settings. Existing frameworks, designed for large academic systems, are often impractical for frontline physicians, creating a dangerous gap between AI adoption and oversight.This article reframes AI governance as a clinician-centered enabler rather than a compliance burden. We propose a pragmatic model built on clear accountability, defined use guardrails, basic safety and bias checks, transparency, and lightweight workflows—supported by a scalable hub-and-spoke approach leveraging trusted professional organizations.Without intentional governance, AI risks amplifying disparities and eroding trust. Done well, it becomes a force multiplier—extending high-quality, equitable care into the communities that need it most. For community physicians, AI governance is not optional; it is essential to protecting patients, preserving clinical judgment, and ensuring that innovation advances equity rather than harm. © 2026 National Medical Association. Published by Elsevier Inc. All rights are reserved, including those for text and data mining, AI training, and similar technologies.}, keywords = {AI governance; Artificial intelligence; Clinical Safety; Community practice; Health equity; Artificial Intelligence; Humans; article; artificial intelligence; clinical practice; clinician; data mining; decision making; health equity; human; medical society; physician; workflow; ethics}, } @article{kamphorstbadewildedelignysferrarifschfermt202659, title = {Opaque procurement: How transparency deficits compromise democratic digital sovereignty}, author = {Kamphorst B.A. and de Wilde de Ligny S. and Ferrari F. and Schäfer M.T.}, year = {2026}, journal = {Journal of Responsible Technology}, volume = {26}, pages = {100178}, doi = {10.1016/j.jrt.2026.100178}, abstract = {While digital sovereignty has quickly become a European policy mantra, few public institutions in Europe possess a clear view of their actual technological dependencies on proprietary infrastructure and non-European providers of data and AI products and services. This lack of insight not only risks turning calls for more sovereignty into hollow slogans but also frustrates the responsible governance of digital technologies. This article argues that opaque procurement practices contribute to the persistence of the problem. Without knowing what is bought, from whom, and under what terms, public institutions are restricted in exercising effective control and cannot adequately plan for and use suitable alternatives. Moreover, without meaningful mechanisms of transparency about procurement, researchers, critical members of the public, and journalists are denied access to necessary resources for democratic contestation and holding public institutions accountable. Building on previous research in the Netherlands, the article uses the Dutch public sector as a case study to show how an opaque procurement ecosystem may undermine democratic digital sovereignty ambitions. To improve the situation, the article outlines strategies for making the procurement of data and AI products and services more transparent. © 2026 The Author(s). Published by Elsevier Ltd. This is an open access article under the CC BY license. http://creativecommons.org/licenses/by/4.0/}, keywords = {AI governance; Data ethics; Digital sovereignty; Procurement; Transparency}, } @article{steinh202660, title = {Toward a taxonomy of generative AI use cases in business contexts: Integrating complexity, risk, and strategy}, author = {Stein H.}, year = {2026}, journal = {Journal of Computational Science}, volume = {96}, pages = {102826}, doi = {10.1016/j.jocs.2026.102826}, abstract = {Enterprises adopting generative AI lack systematic frameworks to classify use cases by complexity, assess associated risks, and sequence implementation according to organizational readiness. We synthesize five perspectives from academic and industry literature — application context, value creation, strategic alignment, technical autonomy, and data governance — to develop a multi-dimensional taxonomy for generative AI deployment. Our taxonomy classifies use cases into four ascending complexity levels: (A) work assistants, (B) automated code generation, (C) system-integrated text generation, and (D) tool use. Each level builds upon prior capabilities while introducing distinct technical, organizational, and risk management requirements. We map these patterns across two application contexts: internal operational efficiency and external customer experience enhancement, showing how risk profiles differ between them. By cross-referencing our taxonomy with the five analytical perspectives, we demonstrate how enterprises can assess current maturity, identify strategically aligned use cases, and construct phased implementation roadmaps that balance innovation velocity with risk governance. This framework bridges technical feasibility assessments with business value realization, enabling evidence-based generative AI adoption across industries. © 2026 The Author. Published by Elsevier B.V. This is an open access article under the CC BY license. http://creativecommons.org/licenses/by/4.0/}, keywords = {Agentic AI; AI governance; Data mesh; Enterprise architecture; Generative AI; Implementation taxonomy; Large language models; Tool use; Risk management; Taxonomies; Agentic AI; AI governance; Application contexts; Data mesh; Enterprise Architecture; Generative AI; Implementation taxonomy; Language model; Large language model; Tool use; Risk assessment}, } @article{kladkoslavazzaayuxfarinam202661, title = {From guidelines to practice: Navigating ethical and cultural challenges in operationalizing AI for project management}, author = {Kladko S. and Lavazza A. and Yu X. and Farina M.}, year = {2026}, journal = {International Journal of Innovation Studies}, volume = {10}, number = {3}, pages = {100181}, doi = {10.1016/j.ijis.2026.100181}, abstract = {The integration of artificial intelligence (AI) into project management is reshaping organizational capabilities by enhancing decision-making, automating routine tasks, and optimizing resource allocation. However, this technological transformation also brings a host of ethical challenges that threaten to undermine trust, transparency, and accountability. Key concerns include algorithmic bias, opacity in decision-making processes, and the erosion of human oversight. Although global ethical frameworks such as those proposed by the EU, OECD, and IEEE offer foundational guidance, they often lack the operational specificity required for application within the dynamic and context-sensitive environment of project management. This paper contends that addressing these shortcomings necessitates a transition from abstract ethical principles to tangible, enforceable mechanisms. Focusing specifically on ethical auditing, this study explores how systematic assessments can be employed to identify, evaluate, and mitigate ethical risks throughout the AI project lifecycle. Drawing upon recent literature and case studies, this paper proposes a multi-dimensional ethical audit model designed for the unique demands of project-based work. By translating normative values into concrete evaluation criteria, ethical audits serve as both diagnostic and preventive tools that can support responsible AI deployment. The paper further emphasizes the critical role of interdisciplinary collaboration in designing audit processes that are context-aware and culturally responsive. © 2026 China Science Publishing and Media Ltd.}, keywords = {AI ethics; Cultural studies; Ethical audits; Project management}, } @article{colapintocgalimberticrepettom202662, title = {Auditing AI systems: Integrating ESG principles for sustainable and ethical AI deployment in the EU}, author = {Colapinto C. and Galimberti C. and Repetto M.}, year = {2026}, journal = {Technovation}, volume = {156}, pages = {103600}, doi = {10.1016/j.technovation.2026.103600}, abstract = {The European Union Artificial Intelligence Act establishes a framework to ensure that artificial intelligence systems deployed within the European Union are safe, compliant with fundamental rights, and aligned with Environmental, Social, and Governance (ESG) principles. However, it lacks specific guidelines to address ESG trade-offs, such as balancing transparency with fairness or environmental concerns with societal impact. This article introduces a novel artificial intelligence audit framework that integrates ESG criteria using the Fuzzy Analytic Hierarchical Approach and Multicriteria Optimization. The framework acts as an institutional translation mechanism and enables decision-makers to prioritize ESG trade-offs and optimally allocate limited resources. By systematically identifying, prioritizing, and addressing the risks of the artificial intelligence system, the framework enhances innovation governance and ensures sustainable and ethical deployment of artificial intelligence. Applying this framework results in improved audit quality, enhanced risk mitigation, and continuous improvement of artificial intelligence systems, thereby promoting positive societal and environmental outcomes. This integrated framework bridges the gap between regulatory expectations and practical implementation, fostering responsible innovation in artificial intelligence in the European Union. Overall, we contribute to the extant literature by repositioning AI governance from a compliance-oriented activity to a dynamic capability that enables firms to manage ESG trade-offs and sustain responsible AI innovation under regulatory uncertainty. © 2026 Elsevier Ltd}, keywords = {Artificial intelligence regulation; ESG criteria in AI; Governance mechanism; Multicriteria optimization; Twin transition; Decision making; Ethical technology; Intelligent systems; International law; International trade; Multiobjective optimization; Regulatory compliance; Sustainable development; AI systems; Artificial intelligence regulation; Artificial intelligence systems; Environmental, social, and governance criteria in AI; European union; Governance mechanisms; Multi-criteria optimisation; Multi-criterion optimization; Trade off; Twin transition; Economic and social effects}, } @article{dockaratrmalhotram202663, title = {Data governance for sustainable AI in organizations: a measurement-oriented capability method with benchmarkability robustness checks and marketplace-microdata operationalization}, author = {Dockara T.R. and Malhotra M.}, year = {2026}, journal = {Discover Artificial Intelligence}, volume = {6}, number = {1}, pages = {574}, doi = {10.1007/s44163-026-01343-8}, abstract = {Sustainable AI governance remains difficult to audit because governance frameworks specify commitments more clearly than evidence thresholds, and sustainability measurement is rarely linked to the operational artifacts that govern AI systems in practice. This paper presents Data Governance for Sustainable AI (DG-SA) as a measurement-oriented capability method rather than a competing normative framework. The method combines four elements: a 12-capability evidence-oriented codebook, a 54-indicator reference assessment instrument, denominator-sensitive benchmarkability ratios, and a marketplace-microdata operationalization that joins external data products to cloud telemetry and disclosure evidence. Empirically, fifteen governance sources are coded against DG-SA using conservative evidence rules. Auditability and human oversight are strongly represented, whereas sustainability-energy and lineage capabilities remain sparse. External benchmarkability is 24.1% across the full instrument, 27.9% across substantive governance items, and 16.7% across thirty core capability indicators; under a direct-only rule the substantive ratio falls to 11.6%. To demonstrate operational usefulness, the study inventories relevant marketplace-distributed datasets across BigQuery sharing, AWS Data Exchange, Snowflake Marketplace, and Databricks Marketplace, and executes a deployment-routing experiment using 44 Google Cloud regions backed by Electricity Maps data. For a fixed 100 kWh workload in 2024, location-based emissions range from 0.273 to 67.876 kg CO2e across available regions, and routing the same batch workload from the median region to the cleanest decile would reduce emissions by 95.1%. The contribution is methodological and integrative rather than universal: DG-SA does not replace principles or management-system standards, but offers a reproducible way to measure, benchmark, and substantiate sustainable AI governance. © The Author(s) 2026.}, keywords = {AI governance; Auditability; Benchmarkability; Carbon intensity; Data governance; Evidence-oriented measurement; Marketplace datasets; Responsible AI; Sustainable AI; Benchmarking; Big data; Commerce; Data Sharing; Government data processing; Sustainable development; AI governance; Auditability; Benchmarkability; Carbon intensity; Data governances; Evidence-oriented measurement; Marketplace dataset; Microdata; Responsible AI; Sustainable AI; Electronic data interchange}, } @article{zaiahadibuzzamanmmcmanusddwalkeya202664, title = {The HALO Model: A Learning Health System Framework for Artificial Intelligence}, author = {Zai A.H. and Adibuzzaman M. and McManus D.D. and Walkey A.}, year = {2026}, journal = {Learning Health Systems}, volume = {10}, number = {3}, pages = {e70100}, doi = {10.1002/lrh2.70100}, abstract = {Introduction: Artificial intelligence is increasingly embedded in healthcare delivery, yet existing Learning Health System (LHS) models do not fully account for the lifecycle management and continuous assurance requirements of AI systems. This gap limits health systems' ability to safely and sustainably integrate AI as a learning component of care. Methods: We conducted a conceptual system modeling investigation grounded in LHS theory and contemporary AI governance frameworks. Through structured theoretical integration, we aligned the classical LHS learning cycle with an action-oriented AI lifecycle and five continuous assurance dimensions, developing a unified framework to support operational implementation within health systems. Results: The resulting Health AI Learning and Oversight (HALO) model specifies how AI functions as a dynamic knowledge artifact within an LHS. Application of the model illustrates how integrating lifecycle stages and continuous assurance instantiates iterative learning loops, enables adaptive governance, and supports operational lifecycle management, including ongoing monitoring of performance, safety, equity, transparency, and security across clinical environments. Conclusions: By extending LHS theory to incorporate AI lifecycle and assurance requirements explicitly, the HALO framework operationalizes continuous learning and oversight for AI-enabled health systems. This model provides a foundation for designing, governing, and sustaining responsible and adaptive AI deployment as healthcare environments evolve. © 2026 The Author(s). Learning Health Systems published by Wiley Periodicals LLC on behalf of University of Michigan.}, keywords = {AI governance; artificial intelligence; learning health system}, } @article{borkowskakjacksond202665, title = {From moral panic to pragmatic governance: reframing AI’s societal impacts in employment, education, and ethics}, author = {Borkowska K. and Jackson D.}, year = {2026}, journal = {AI and Society}, volume = {41}, number = {5}, pages = {5371-5383}, doi = {10.1007/s00146-026-02921-1}, abstract = {Public debate about AI shifts between existential alarm and complacent reassurance. We read this volatility through the classic lens of moral panic—a discourse that amplifies diffuse fears and pulls policy toward symbolic gestures—and set it against a pragmatic program that treats risks as diagnosable and governable. Reconstructing AI’s shift from task-specific tools to an inference infrastructure clarifies why anxiety spikes: authority has moved upstream to data curators, model/compute providers, and benchmark communities, while outputs grow more fluent and more opaque. Rather than adjudicating the panic, we translate it into action. This paper operationalizes pragmatism by defining the actionable pathway: measurable controls, independent oversight, and cross-sector coordination that keep AI’s promised gains while tightening accountability where it matters. Its portability is demonstrated across employment, education, and ethics. Then drawing on a Neo-Triple Helix perspective, we propose a governance framework that coordinates universities, industry, government, and civil society through participation mechanisms, assurance guarantees, and procurement levers. This approach aims to translate principles into verifiable practice, grounding AI governance in evidence, inclusivity, and transparent accountability to guide AI development toward broad societal benefit. © The Author(s) 2026.}, keywords = {Human–AI teaming; Moral panic; Neo-Triple Helix (NTH); Participatory risk-based regulation; Pragmatic risk governance; Artificial intelligence; Ethical technology; Public policy; Human–AI teaming; Moral panic; Neo-triple helix; Participatory risk-based regulation; Pragmatic risk governance; Reframing; Risk governance; Risk-based regulation; Societal impacts; Triple helixes; Risk management}, } @article{liumalmugrenichotiavsahorenkurucza202666, title = {Experiential learning and governance in the socio-technical era: Modeling responsible AI performance via explainability and adaptability}, author = {Liu M. and Almugren I. and Chotia V. and Sahore N. and Kurucz A.}, year = {2026}, journal = {Technological Forecasting and Social Change}, volume = {227}, pages = {124624}, doi = {10.1016/j.techfore.2026.124624}, abstract = {The concept of artificial intelligence (AI) is altering the way organizations operate. AI systems will deliver more intelligent results in a shorter period of time, starting with decision-making up to innovation. However, the more it is adopted, the more issues to do with fairness, transparency, and accountability are raised. Most organizations are finding it difficult to reconcile innovation and ethical responsibility. This study discusses the role of internal capabilities in making firms govern AI responsibly. The study proposes a model linking four key organizational capabilities, i.e., explainable AI capability, contextual learning adaptability, experiential learning orientation, and organizational ethical alignment to responsible AI performance. The impact of these capabilities on user interpretability and trust, responsible AI governance maturity, and decision transparency is also examined in this study. The results show that explainable AI capability and learning adaptability enhance user trust, while experiential learning orientation and organizational ethical alignment significantly improve governance maturity. Governance maturity and decision transparency lead to stronger responsible AI performance. Interestingly, not all expected paths held as user interpretability trust and governance maturity did not directly predict decision transparency. The findings show that building technical and cultural capabilities inside firms is essential not just to deploy AI effectively, but to do it responsibly. For leaders, this means moving beyond checklists and toward meaningful governance rooted in learning, transparency, and ethical alignment. © 2026 Elsevier Inc.}, keywords = {Contextual learning adaptability; Decision transparency; Experiential learning orientation; Explainable AI capability; Organizational ethical alignment; Responsible AI governance maturity; Responsible AI performance; User interpretability trust; Alignment; Artificial intelligence; Decision making; Ethical technology; Learning systems; Contextual learning; Contextual learning adaptability; Decision transparency; Experiential learning; Experiential learning orientation; Explainable artificial intelligence capability; Interpretability; Learning orientation; Organisational; Organizational ethical alignment; Performance; Responsible artificial intelligence governance maturity; Responsible artificial intelligence performance; User interpretability trust; artificial intelligence; ethics; governance approach; learning; performance assessment; Transparency}, } @article{elsayedn202667, title = {Clinical informatics governance framework for scalable and explainable AI-enabled assistive systems in mental health care}, author = {Elsayed N.}, year = {2026}, journal = {Health Policy and Technology}, volume = {15}, number = {6}, pages = {101242}, doi = {10.1016/j.hlpt.2026.101242}, abstract = {Objectives: Artificial intelligence (AI)-enabled assistive systems, including conversational agents and socially assistive robots, are increasingly proposed to address gaps in mental health services. However, large-scale deployment remains constrained by fragmented governance, unclear decision logic for embodiment, inconsistent explainability standards, and limited integration into health system workflows. This study develops a policy-relevant governance framework to support scalable and accountable deployment of AI-enabled assistive systems in mental health care. Methods: We conducted a structured conceptual synthesis of literature in clinical informatics, AI governance, risk-based regulation, implementation science, and digital mental health. Drawing on these domains, we developed a cyclical governance framework that links task risk stratification, embodiment modality selection, risk-proportionate explainability, and workflow-aligned evaluation within a learning health system. Results: The framework introduces a risk-tiered governance architecture in which oversight intensity and explainability depth scale with task risk. A virtual-first principle guides embodiment decisions, balancing clinical benefit with operational and regulatory burden. We further propose a Relative Governance–Operational Complexity Index (RGOCI) to support policy-level decision-making regarding embodiment trade-offs and sustainability. The framework positions explainability as governance infrastructure rather than a technical add-on, enabling accountability and regulatory alignment. Conclusions: Scalable AI deployment in mental health requires adaptive governance structures that integrate risk stratification, accountability mechanisms, and realities of health system implementation. The proposed framework provides policy-oriented architecture to guide responsible digital mental health innovation and regulatory calibration. Public Interest Abstract: Artificial intelligence tools such as mental health chatbots and assistive robots are increasingly promoted as solutions to shortages in mental health services. However, simply introducing these technologies does not guarantee safe or sustainable use. Without clear governance, oversight, and accountability structures, digital tools may create unintended risks or fail to integrate into routine care. This article proposes a practical governance framework to help health systems and policymakers decide how AI tools should be deployed responsibly. The framework emphasizes matching oversight and transparency to clinical risk, prioritizing simpler virtual systems when appropriate, and continuously evaluating real-world performance. By providing structured guidance for regulators, health system leaders, and technology developers, this work aims to support safe, equitable, and sustainable expansion of AI-enabled mental health services. © 2026 Fellowship of Postgraduate Medicine}, keywords = {AI governance; Artificial intelligence in healthcare; Digital mental health; Explainable artificial intelligence; Learning health systems; Risk-based regulation; Article; decision making; digital health; explainable artificial intelligence; health care policy; human; implementation science; learning health system; medical informatics; mental health care; risk assessment; workflow}, } @article{seosparkjileemyoosleeukkimc202668, title = {Uncovering Hierarchical Asymmetries in Artificial Intelligence Transformation: Navigating the Bright and Dark Sides Across Organizational Levels}, author = {Seo S. and Park J.-I. and Lee M. and Yoo S. and Lee U.-K. and Kim C.}, year = {2026}, journal = {Journal of Visualized Experiments}, volume = {2026-May}, number = {231}, pages = {e70756}, doi = {10.3791/70756}, abstract = {Governance structures that align executive strategy with frontline implementation depend on understanding how organizational hierarchy shapes perceptions of artificial intelligence (AI) transformation (AX) success factors. We present a validated, replicable survey-based methodology coupling partial least squares structural equation modeling (PLS-SEM) with multi-group analysis (MGA) to detect and quantify hierarchical asymmetries in AX perceptions—going beyond aggregate analysis to expose structural divides that single-group methods leave hidden. Drawing on 293 professionals involved in AI projects, we assess four dynamic capability dimensions —operational agility, data readiness, customer proximity, and strategic process discipline—together with key enabling factors: technological support, AI-sensitive risk tolerance, environmental context, and proactive leadership. Compositional invariance is first verified through MICOM testing; only after confirming between-group comparability does the analysis move to permutation-based MGA significance tests. Both groups exhibit positive associations between dynamic capabilities and AX performance (executives: β = 0.637, R2 = 0.406; practitioners: β = 0.531, R2 = 0.282). Permutation-based MGA detects no statistically significant differences in structural path coefficients between the two groups (all p ≥ .178); the hypothesized moderating role of organizational position on path relationships (H5) therefore lacks support. Yet MICOM Step 3a shows executives rating all major constructs significantly higher than practitioners do (8 of 11 constructs, all p ≤ .045)—evidence that hierarchical asymmetry surfaces as a gap in perceptual levels, not in the structural relationships themselves. As a transferable methodological template, the protocol equips researchers studying perception gaps in organizational transformation with tools bearing directly on AI governance, change management strategy, and cross-level alignment in technology-intensive settings. © 2026, Journal of Visualized Experiments Journal of Visualized Experiments.}, } @article{ngaicsbchenrswangyjinx202669, title = {Crisis Communication Readiness in the Age of Digital Transformation and GenAI: Insights From Hong Kong}, author = {Ngai C.S.B. and Chen R.S. and Wang Y. and Jin X.}, year = {2026}, journal = {Journal of Contingencies and Crisis Management}, volume = {34}, number = {2}, pages = {e70159}, doi = {10.1111/1468-5973.70159}, abstract = {This study investigates the impact of digital transformation and emerging technologies, particularly generative artificial intelligence (GenAI), on the readiness and practices of crisis communication professionals in Hong Kong (HK). Despite growing theoretical advances, recent literature remains predominantly Western-centric, with limited empirical research examining how Asia-Pacific professionals navigate digital disruption standing for an important gap to be addressed given the region's rapid digital adoption and contextual sensitivity. Through qualitative content analysis, the research identifies key themes related to organizational adaptation in a rapidly evolving informational landscape in HK. Findings reveal that the integration of AI-driven tools and advanced analytics has significantly enhanced the organizational readiness mindset and efficiency. However, the accelerated flow of digital information also introduces new challenges, emphasizing the indispensable role of critical human judgement while recommending a hybrid human–AI collaboration approach for effective crisis communication and management. This study further highlights the importance of cultivating digital fluency, core skillsets, strategic stakeholder coordination, and developing a human-centric AI governance framework for crisis communication professionals navigating complex crises proactively and resiliently. By addressing notable empirical and theoretical gaps, this research offers important theoretical insights into readiness applicability in the HK context, while providing organizations and practitioners with actionable strategies to enhance their crisis readiness in the digital era. © 2026 The Author(s). Journal of Contingencies and Crisis Management published by John Wiley & Sons Ltd.}, keywords = {artificial intelligence; crisis communication; readiness; China; Hong Kong; artificial intelligence; communication; crisis management; technological change}, } @article{mohammedbnourddineeguerssfz202670, title = {Artificial intelligence driven predictive governance of stadium security in Morocco, France and Qatar ahead of AFCON 2025 and the 2030 FIFA World Cup}, author = {Mohammed B. and Nourddine E. and Guerss F.-Z.}, year = {2026}, journal = {Discover Artificial Intelligence}, volume = {6}, number = {1}, pages = {554}, doi = {10.1007/s44163-026-01274-4}, abstract = {Stadium security is increasingly shaped by the introduction of artificial intelligence (AI), yet little is known about how national governance structures influence its deployment. This study compares Morocco, France, and Qatar to examine how AI-assisted security systems are framed, regulated, and operationally integrated in three contrasting contexts. Using a purposive sample of 22 institutional and regulatory documents and 34 scientific sources, the analysis applies a seven-dimension framework covering legal foundations, technological capacity, institutional coordination, staff training, ethical safeguards, predictive mechanisms, and post-event learning. The results indicate substantial cross-national divergence in governance maturity and operational integration patterns. France benefits from a dense legal framework and strong ethical oversight, which both constrain and stabilize AI adoption. Qatar’s model reflects the specific pressures of hosting the FIFA World Cup 2022, leading to the rapid deployment of advanced, centralized monitoring infrastructures. Morocco presents an evolving configuration: legal foundations and institutional bodies are in place, but technological integration and coordination remain uneven across regions. Across all cases, the study identifies a persistent gap between the growing availability of AI tools and their use for long-term organizational learning. The findings highlight that AI-enabled stadium security depends less on technology itself than on the institutional environment that governs its use. As Morocco prepares for CAN 2025 and the 2030 World Cup, strengthening coordination, clarifying regulatory boundaries, enhancing training, and formalizing review processes appear essential. Beyond its empirical contribution, the study offers a conceptual framework that supports cross-national analysis of AI governance in large public venues. © The Author(s) 2026.}, keywords = {Artificial intelligence (AI); Comparative policy analysis; Morocco–France–Qatar case study; Predictive safety systems; Risk and crowd management; Stadium security governance; Surveillance technologies; Artificial intelligence; Human resource management; Personnel training; Philosophical aspects; Predictive analytics; Public policy; Risk assessment; Risk management; Safety engineering; Security systems; Stadiums; Artificial intelligence; Case-studies; Comparative policy analyze; Crowd managements; Morocco–france–qatar case study; Policy analysis; Predictive safety system; Risks management; Security governances; Stadium securities; Stadium security governance; Surveillance technology; Recreation centers}, } @article{slatterypsaeriakgrundyeacgrahamjnoetelmuukrdaojpourscaspersthompsonn202671, title = {The AI risk repository: A meta-review, database, and taxonomy of risks from artificial intelligence}, author = {Slattery P. and Saeri A.K. and Grundy E.A.C. and Graham J. and Noetel M. and Uuk R. and Dao J. and Pour S. and Casper S. and Thompson N.}, year = {2026}, journal = {Patterns}, volume = {7}, number = {5}, pages = {101517}, doi = {10.1016/j.patter.2026.101517}, abstract = {The risks posed by artificial intelligence (AI) concern academics, auditors, policymakers, AI companies, and the public. Researchers, policymakers, and technology companies discuss AI risks using inconsistent terminology—the same word may describe different problems, while different words describe identical concerns. This fragmentation impedes coordinated responses to AI challenges. We address this by creating the AI Risk Repository: a living database of 1,725 risks extracted from 74 existing taxonomies and frameworks. We organize these risks using two complementary classification systems. The Causal Taxonomy classifies risks by their origins: which entity causes them (human or AI), whether intentional, and when they occur (before or after deployment). The Domain Taxonomy classifies risks by their effects across seven areas, from discrimination and privacy violations to misinformation and weapons development. This shared reference enables more coordinated approaches to discussing, researching, auditing, and governing AI systems across sectors and jurisdictions. © 2026 The Authors.}, keywords = {AI governance; AI safety; machine-learning governance; risk assessment; societal impact of AI; Artificial intelligence; Copyrights; Learning systems; Public risks; Risk perception; Safety engineering; Taxonomies; Artificial intelligence governance; Artificial intelligence safety; Classification system; Machine-learning; Machine-learning governance; Policy makers; Risks assessments; Societal impact of artificial intelligence; Societal impacts; Technology companies; Risk assessment}, } @article{shiraishik202672, title = {Who owns meaning in an age of AI? Beyond transparent systems to shared cosmologies}, author = {Shiraishi K.}, year = {2026}, journal = {AI and Society}, volume = {41}, number = {5}, pages = {5363-5369}, doi = {10.1007/s00146-026-02925-x}, abstract = {AI governance in healthcare increasingly foregrounds transparency, auditability, and traceability. Yet what these frameworks often secure is a surface-level domain of factual correctness and accountable documentation—what this paper terms the semantic meaning layer (SML). In clinical practice, however, trust and ethical acceptability are also anchored in deeper, slower-moving structures: cultural cosmologies, value commitments, embodied practices, and lived experience—what this paper terms the cosmological meaning layer (CML). A key risk in contemporary deployment is, therefore, not only “factual hallucination” but “meaning hallucination”: outputs that are technically coherent yet ethically hollow, socially misaligned, or experientially discordant. The paper proposes the SML–CML framework as a co-constituted stratification: analytically distinguishable but not ontologically separable. Drawing on abductive clinical reasoning—especially traditional Chinese medicine (TCM) pattern differentiation—we show how diagnosis is often an act of meaning-construction shaped by embodied perception (touch, trained attention) and shared worldviews, not merely an inferential mapping from data to labels. We then clarify “whose trust” transparency regimes often secure, distinguishing relational trust (patient–clinician) from institutional legitimacy (audit, regulation, medico-legal protection). Finally, we show how global coding infrastructures (e.g., ICD-11) and AI evaluation practices can strengthen comparability while structurally excluding local cosmologies and embodied meaning. We argue that preventing AI-induced hollowing of meaning requires governance that explicitly addresses worldview pluralism and supports “shared abduction”: collaborative, value-explicit inquiry in which humans and institutions negotiate not only correctness but also the cosmologies that make care meaningful. © The Author(s), under exclusive licence to Springer-Verlag London Ltd., part of Springer Nature 2026.}, keywords = {Abduction; AI governance; Cosmotechnics; Decolonial AI; ICD-11; Narrative medicine; Traditional Chinese medicine; Trust; Diagnosis; Ethical aspects; Laws and legislation; Semantics; Transparency; Abduction; AI governance; Clinical practices; Cosmotechnic; Decolonial AI; ICD-11; Moving structures; Narrative medicine; Traditional Chinese Medicine; Trust; Cosmology}, } @article{floridilbuttabonicgertlernhineemorleyjnovellicschrodert202673, title = {Agentic AI Optimisation (AAIO): What It Is, How It Works, Why It Matters, and How to Deal with It}, author = {Floridi L. and Buttaboni C. and Gertler N. and Hine E. and Morley J. and Novelli C. and Schroder T.}, year = {2026}, journal = {Minds and Machines}, volume = {36}, number = {2}, pages = {25}, doi = {10.1007/s11023-026-09779-8}, abstract = {The emergence of Agentic Artificial Intelligence (AAI) systems capable of independently initiating digital interactions requires a new optimisation paradigm designed explicitly for seamless agent-platform interactions. This article introduces Agentic AI Optimisation (AAIO) as an essential methodology for ensuring effective integration between websites and agentic AI systems. As Search Engine Optimisation (SEO) has shaped digital content discoverability, AAIO can define interactions between autonomous AI agents and online platforms. By examining the mutual interdependence between website optimisation and agentic AI success, the article highlights the virtuous cycle that AAIO can create. It further explores the governance, ethical, legal, and social implications (GELSI) of AAIO, emphasising the need for proactive regulatory frameworks to mitigate potential negative impacts. The article concludes by affirming AAIO’s essential role as part of a fundamental digital infrastructure in the era of autonomous digital agents, advocating equitable and inclusive access to its benefits. © The Author(s), under exclusive licence to Springer Nature B.V. 2026.}, keywords = {Agentic AI Optimisation (AAIO); Agentic Artificial Intelligence (AAI); AI ethics; Digital governance; Digital optimisation; Artificial intelligence; Electronic document exchange; Ethical aspects; Government data processing; Intelligent agents; Laws and legislation; Search engines; Websites; Agent platform; Agentic AI optimization; Agentic artificial intelligence; AI ethic; AI systems; Artificial intelligence systems; Digital governance; Digital interactions; Digital optimization; Optimisations; Autonomous agents}, } @article{defossezdnapierl202674, title = {Is the EU AI Act future proof?}, author = {Defossez D. and Napier L.}, year = {2026}, journal = {Revista de Direito, Estado e Telecomunicacoes}, volume = {18}, number = {2}, pages = {444-477}, doi = {10.26512/lstr.v18i2.62431}, abstract = {[Purpose] The EU AI Act, adopted in March 2024 and largely effective by 2026, represents the world's first comprehensive horizontal AI regulation. Building on the EU's experience with the GDPR, it employs a risk-based approach, broad definitions for AI systems and General-Purpose AI Models (GPAI) and asserts extensive extraterritorial reach. This article critically examines whether the AI Act is truly “future-proof” against AI’s rapid evolution. [Methodology/approach/design] The article employs doctrinal legal analysis of the EU AI Act's provisions, particularly its risk-based classification framework, examining the Act's structural architecture against established risk governance principles and comparing its approach with emerging national legislation and international frameworks. The analysis focuses on the interaction between Articles 5, 6, 9, 51, and 55, examining how static regulatory classifications interact with dynamic AI capabilities. [Findings] While the Act demonstrates commendable ambition- including technology-neutral definitions, extraterritorial reach, integration with existing frameworks like the GDPR, and establishment of a centralized AI Office- it suffers from a fundamental architectural flaw that compromises its long-term viability. Key weaknesses include a highly complex and potentially slow enforcement mechanism, risks of EU consumer discrimination due to compliance burdens, and the potential for regulatory circumvention. A more fundamental flaw is the regulation of inherently dynamic AI systems through static risk classifications. This static approach creates three cascading failures. First, it cannot capture AI systems that evolve post-deployment from low-risk to high-risk. Second, rigid thresholds (particularly the 10^25 FLOPs threshold for systemic risk GPAI) incentivize regulatory arbitrage. Third, enforcement mechanisms remain reactive rather than proactive, creating a ‘regulatory firefighting’ approach incompatible with genuine future-proofing. While Article 9's risk management framework is theoretically sound, it operates within this static classification system and cannot compensate for the broader structural weakness. [Practical implications] The Act represents a vital first step in AI governance, but its static classification, internal inconsistencies, and unaddressed challenges severely compromise its long-term adaptability- means it will require continuous reactive amendments to remain relevant. It is less a future-proof edifice than a foundational structure already showing cracks. The Act's static architecture is already prompting Member States to develop conflicting supplementary regulations (exemplified by Italy's AI Law), creating fragmentation rather than harmonization. © 2026 Universidade de Brasilia. All rights reserved.}, keywords = {Artificial intelligence regulation; Dynamic systems; EU AI Act; Future-proof legislation; Risk-based approach}, } @article{bohrj202675, title = {Coordination transparency: governing distributed agency in AI systems}, author = {Bohr J.}, year = {2026}, journal = {AI and Society}, volume = {41}, number = {5}, pages = {4767-4777}, doi = {10.1007/s00146-026-02853-w}, abstract = {AI governance frameworks designed for human decision-making fail when consequential outcomes emerge from coordination among machines. Current approaches create governance illusions—interfaces suggest control while algorithmic coordination unfolds beyond effective intervention. This article develops coordination transparency as a governance mechanism grounded in sociomaterial accounts of distributed agency. Instead of restoring centralized human control through better interfaces, coordination transparency targets agent-to-agent interactions directly through four components: interaction logging, live coordination monitoring, intervention hooks, and boundary conditions. This approach resolves a category error in prevailing frameworks that apply human-centered tools to distributed coordination processes. The framework shifts oversight from post hoc explanation of individual outputs to real-time observation and steering of coordination patterns where behavior actually emerges, preserving democratic accountability in systems characterized by distributed rather than centralized agency. © The Author(s) 2026.}, keywords = {AI governance; Algorithmic coordination; Coordination transparency; Distributed agency; Multi-agent systems; Sociomaterial governance; Association reactions; Behavioral research; Computational methods; Decision making; Information systems; Intelligent agents; Man machine systems; Real time systems; Transparency; AI governance; AI systems; Algorithmic coordination; Algorithmics; Centralised; Coordination transparency; Distributed agencies; Human decision-making; Multiagent systems (MASs); Sociomaterial governance; Multi agent systems}, } @article{aveggioccartersrmukundanhcameronebeguerras202676, title = {Aligning innovation and security in AI-enabled biotechnology: a framework for designing and funding mutually reinforcing approaches}, author = {Aveggio C. and Carter S.R. and Mukundan H. and Cameron E.B.E. and Guerra S.}, year = {2026}, journal = {Frontiers in Microbiology}, volume = {17}, pages = {1820739}, doi = {10.3389/fmicb.2026.1820739}, abstract = {The convergence of artificial intelligence and biotechnology is transforming the life sciences and enabling rapid design, development, and analysis across the research lifecycle. However, this acceleration also heightens the risk for potential biological misuse concerns. Building a sustainable and secure bioeconomy requires moving beyond rhetoric about balancing innovation and security toward practical, operational efforts that do both. This paper proposes a framework for initiatives that embed safety and trustworthiness into the architecture of biological research systems while simultaneously advancing scientific progress. Alongside this framework, we will describe three case studies where investment can yield mutual benefit: (1) secure and tiered data-sharing infrastructures that broaden access to high-quality biological data while maintaining control over sensitive information; (2) provenance and metadata tracking mechanisms for AI enabled biological tools that enhance scientific reproducibility and oversight; and (3) capability benchmarking approaches for AI-enabled biological tools that can enable performance improvements while providing situational awareness about biological risk trajectories. Our framework demonstrates how co-designing innovation and security objectives can transform potential trade-offs into reinforcing outcomes. The paper concludes by outlining policy and funding strategies to enable such mutually reinforcing win-win approaches, positioning responsible AI-enabled biotechnology as both a driver of innovation and a foundation for global biosecurity. Copyright © 2026 Aveggio, Carter, Mukundan, Cameron and Guerra.}, keywords = {AI-enabled biological tools; artificial intelligence; biosecurity governance; biotechnology; dual-use governance; acceleration; article; artificial intelligence; awareness; benchmarking; biosecurity; biotechnology; controlled study; funding; human; infrastructure; investment; metadata; reproducibility; responsible artificial intelligence; security; trustworthiness}, } @article{songb202677, title = {Algorithmic bias in HR recruitment systems: A qualitative analysis of managerial risk and sociological implications}, author = {Song B.}, year = {2026}, journal = {PLOS ONE}, volume = {21}, number = {6 June}, pages = {e0349400}, doi = {10.1371/journal.pone.0349400}, abstract = {Algorithmic hiring systems have rapidly proliferated across industries, promising improved efficiency, objectivity, and scalability in recruitment processes. However, growing empirical evidence reveals a gap between these expected benefits and actual outcomes, as many systems inadvertently reproduce or amplify historical inequalities embedded in training data. The main aim of this study is to develop and evaluate a rigorous, multi-layered framework capable of identifying, interpreting, and mitigating bias throughout the full lifecycle of algorithmic hiring systems, ensuring both immediate decision fairness and long-term career equity. To achieve this aim, the Multi-Layer Bias Analysis and Mitigation System (ML-BAMS) is introduced as a comprehensive approach to detecting and mitigating bias in HR recruitment systems. The framework integrates modules for bias decomposition and data diagnostics, recruitment-aware fairness evaluation across multi-stage pipelines, interpretability of opaque models through influence mapping, fairness-preserving representation learning, and longitudinal simulation of career mobility outcomes. Using synthetic hiring datasets, the proposed framework demonstrates substantial reductions in demographic disparities across key fairness metrics, including improvements in demographic parity (32.4%), equal opportunity (28.7%), equalized odds (25.9%), and treatment equality (19.2%), while maintaining competitive predictive accuracy (ΔAccuracy: + 0.024). These findings highlight the importance of integrated sociotechnical approaches that address bias transmission, enhance transparency, and account for long-term impacts. The ML-BAMS framework provides a practical and modular toolset for implementing responsible AI in recruitment, balancing operational performance with ethical considerations of fairness and social equity. © 2026 Baijia Song. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.}, keywords = {Algorithms; Bias; Career Mobility; Humans; Personnel Selection; algorithm bias; article; benchmarking; career mobility; decomposition; diagnosis; fairness; feature learning (machine learning); human; machine learning; open access publishing; parity; qualitative analysis; responsible artificial intelligence; simulation; social equity; algorithm; personnel management; procedures; statistical bias}, } @article{hanmzhaoj202678, title = {The rise and fall of AI's benefits: Investigating the curvilinear relationship between employee-AI collaboration and successful aging at work}, author = {Han M. and Zhao J.}, year = {2026}, journal = {Journal of Vocational Behavior}, volume = {168}, pages = {104265}, doi = {10.1016/j.jvb.2026.104265}, abstract = {The rapid integration of artificial intelligence (AI) into workplaces presents both opportunities and challenges for aging workforces, yet existing research lacks a nuanced understanding of how employee-AI collaboration influences successful aging at work. Drawing on conservation of resources theory, we theorize that employee-AI collaboration has an inverted U-shaped relationship with successful aging at work, mediated by job crafting towards strengths. We further identify AI ethical anxiety as a critical boundary condition that intensifies this curvilinear relationship. Data collected from 312 older employees across three waves supported the inverted U-shaped moderated mediation model. Specifically, we found that increases in employee-AI collaboration are associated with higher job crafting towards strengths and more successful aging at work up to an inflection point, beyond which further employee-AI collaboration yields detrimental effects. AI ethical anxiety accelerates the decline of employee-AI collaboration's positive effects while exacerbating its negative consequences. Our findings challenge prevailing linear assumptions about the influences of employee-AI collaboration and highlight the need for a balanced and ethically conscious AI implementation that safeguards older employees' career development. © (2026), (Academic Press Inc.). All rights reserved.}, keywords = {AI ethical anxiety; Employee-AI collaboration; Job crafting towards strengths; Successful aging at work}, } @article{raptisapapadakisn202679, title = {Securitizing the algorithm: a quantitative analysis of liberalization and securitization logics in the EU Artificial Intelligence Act}, author = {Raptis A. and Papadakis N.}, year = {2026}, journal = {Frontiers in Political Science}, volume = {8}, pages = {1806998}, doi = {10.3389/fpos.2026.1806998}, abstract = {Introduction – The EU Artificial Intelligence Act (AI Act) represents a global landmark in digital governance, yet the underlying political rationale structuring its provisions—securitization versus liberalization—remains underexplored. This study examines whether the AI Act constructs AI as an existential threat requiring exceptional measures or embeds it within the Union’s liberal regulatory tradition, while also developing a replicable framework for analyzing securitization in regulatory texts. Methods – The study introduces the Securitization–Liberalization Balance Index (SLBI), a quantitative measurement framework that operationalizes securitization theory into structured indicators and applies them systematically to the full text of the AI Act. Results – The findings reveal a hybrid regulatory configuration. Securitizing elements are concentrated in high-risk domains and prohibited practices, while liberalizing provisions display a consistent predominance in both average intensity and cumulative weight. At the same time, the SLBI demonstrates high internal consistency and analytical coherence, capturing structured variation across articles and chapters and supporting its robustness as a measurement instrument. Discussion – The AI Act reflects a form of regulated and bounded securitization, in which security concerns are normalized through ordinary legal and governance mechanisms rather than emergency politics. This trajectory is further reinforced by recent EU initiatives, such as the “Digital Omnibus, ” which emphasize regulatory consolidation and simplification rather than the expansion of securitization dynamics. The study proposes the SLBI as a replicable methodological framework for the empirical assessment of securitization dynamics in regulatory texts, contributing to the debate on the meta-security era. Copyright © 2026 Raptis and Papadakis.}, keywords = {AI governance; artificial intelligence regulation; digital governance; Digital Omnibus; EU Artificial Intelligence Act (AI Act); liberalization; securitization theory; Securitization–Liberalization Balance Index (SLBI)}, } @article{fangszhuxzhutzhanglliy202680, title = {Fair enough? The unintended consequences of service robot social presence for customer fairness perception in hospitality industry}, author = {Fang S. and Zhu X. and Zhu T. and Zhang L. and Li Y.}, year = {2026}, journal = {International Journal of Hospitality Management}, volume = {138}, pages = {104730}, doi = {10.1016/j.ijhm.2026.104730}, abstract = {The growing prevalence of service robots in the hospitality industry has exposed heightened ethical concerns, particularly regarding fairness considerations in human-robot interactions. The responsible and ethical application of AI has emerged as an urgent priority. This research utilizes five experiments to investigate what, how and when service robots impact customer fairness perception in hospitality industry. The results indicate that service robots’ social presence negatively influences customer fairness perception. Customers' mind perception partially mediates this effect, specifically through their perception of the robot experience rather than perceived agency. Furthermore, service robot's role as an augmentation (vs. substitution) moderates the effects, with a stronger positive impact in the substitution condition. This research mainly contributes to the literature on responsible AI by exploring service robots’ social presence in influencing customer fairness perceptions. The findings provide hospitality practitioners on how to effectively integrate service robots to enhance perceived responsible AI and improve the overall customer experience. © 2026 Elsevier Ltd}, keywords = {Customer fairness perception; Hospitality industry; Mind perception; Responsible AI; Service robot; Social presence}, } @article{joharib202681, title = {Dual erosion of equality and remedy in algorithmic decision systems}, author = {Johari B.}, year = {2026}, journal = {Discover Artificial Intelligence}, volume = {6}, number = {1}, pages = {245}, doi = {10.1007/s44163-026-01040-6}, abstract = {This article develops a dual erosion framework demonstrating how Artificial Intelligence systems simultaneously undermine equality rights through discriminatory outcomes and obstruct effective remedies through opacity. Through comparative case synthesis spanning criminal justice algorithms in the United States and Canada, welfare eligibility systems in the European Union (EU) and India, credit scoring in Germany, employment screening, and content moderation, the analysis reveals that equality risk and remedy risk operate interdependently rather than in parallel. Biased algorithmic outputs become unreviewable due to opacity, while opacity enables bias to persist undetected. The article examines regulatory responses under the EU Artificial Intelligence Act, the Digital Services Act, and the revised Santa Clara Principles, proposing concrete implementation frameworks that include discovery protocols reconciling trade secrets with due process, burden-shifting standards adapted from employment discrimination law, risk-tiered due diligence requirements, independent audit mechanisms, and participatory governance models. These findings advance algorithmic accountability scholarship by providing adjudicable standards for courts and regulators, while demonstrating that comprehensive governance requires integrated approaches that address both equality and remedy dimensions. © The Author(s) 2026.}, keywords = {Algorithmic accountability; Algorithmic transparency; Artificial intelligence governance; Digital rights regulation; Discrimination law; Due process rights; Effective remedy; Artificial intelligence; Copyrights; Discriminators; Erosion; Government data processing; International law; Opacity; Algorithmic accountability; Algorithmic transparency; Algorithmics; Artificial intelligence governance; Digital right regulation; Digital rights; Discrimination law; Due process right; Effective remedy; Risk assessment}, } @article{alhefaitysrsamohamadejamlimritotlarasatiamohamadna202682, title = {Trust in government as a mediator in the relationship between AI implementation and citizen satisfaction: Evidence from UAE Policing}, author = {Alhefaity S.R.S.A. and Mohamad E. and Jamli M.R. and Ito T. and Larasati A. and Mohamad N.A.}, year = {2026}, journal = {Multidisciplinary Science Journal}, volume = {8}, number = {5}, pages = {e2026292}, doi = {10.31893/multiscience.2026292}, abstract = {This study investigates the influence of Artificial Intelligence (AI) implementation on citizen satisfaction with police services in the United Arab Emirates (UAE), with a specific focus on the mediating role of trust in government. As smart policing initiatives expand under the UAE’s digital governance agenda, understanding how AI technologies shape public trust and service perceptions becomes increasingly critical for sustainable public engagement. A quantitative research design was adopted using a cross-sectional survey method. Data were collected from 365 police personnel within the Abu Dhabi Police across AI-enhanced, operational, and administrative units. Structural Equation Modeling (PLS-SEM) was employed to assess both direct and indirect relationships among AI implementation, trust in government, and citizen satisfaction. Validated scales and reliability checkswere used to ensure measurement accuracy. The results demonstrate that AI implementation has a significant positive effect on citizen satisfaction (β = 0.400, p < 0.001) and trust in government (β = 0.511, p < 0.001). Trust in government, in turn, significantly influences citizen satisfaction (β = 0.321, p < 0.001) and mediates the relationship between AI implementation and satisfaction (indirect effect β = 0.164, p < 0.001). These findings support the conceptual framework grounded in Public Value Theory and Expectation-Confirmation Theory, highlighting the importance of trust as a critical mechanism through which AI influences public attitudes toward policing services. This study offers actionable guidance for policymakers and law enforcement agencies seeking to integrate AI responsibly. Emphasis should be placed on transparency, citizen-centric design, and ethical AI governance to reinforce trust and ensure positive service experiences. AI strategies must be aligned with public expectations to sustain long-term satisfaction. The study contributes to emerging literature on digital public service delivery by empirically validating the mediating role of trust in government in AI-supported law enforcement. It is among the first to explore this dynamic in the context of UAE policing, offering region-specific insights into the interplay between technology, institutional trust, and citizen satisfaction. Copyright (c) 2025 The Authors. This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.}, keywords = {artificial intelligence; citizen satisfaction; trust in government; United Arab Emirates}, } @article{shilazianfumtahirz202683, title = {Participatory urban innovation: how community creativity drives sustainable city development in the digital era}, author = {Shi L. and Azian F.U.M. and Tahir Z.}, year = {2026}, journal = {SN Business and Economics}, volume = {6}, number = {5}, pages = {127}, doi = {10.1007/s43546-026-01138-0}, abstract = {Cities are under growing pressure to become more sustainable while also involving residents in shaping urban futures. This study examines how participatory urban innovation relates to sustainable development in the digital era and whether digital inclusion and city context shape this relationship. Using a panel of 100 medium and large cities observed from 2015 to 2024, we construct a participatory innovation index, a multidimensional sustainable development index, and a digital inclusion index. We estimate city–year fixed-effects models with robustness checks, lagged specifications, placebo tests, heterogeneity analysis, and a simple mediation framework. The results show that higher levels of participatory innovation are associated with small but non-trivial improvements in sustainable development performance, even after controlling for income, digital infrastructure, environmental regulation, and time-invariant city characteristics. This study contributes to the discourse on AI ethics and social intelligence by examining how participatory innovation and digital inclusion shape sustainable urban development within the broader context of digital transformation. Digital inclusion is positively related to sustainable development and reduces the size of the participatory innovation coefficient, although evidence for strong mediation is limited. The association between participatory innovation and sustainability is present in both emerging and more developed cities and appears somewhat stronger in emerging contexts. This study highlights the ethical implications of artificial intelligence in the context of digital transformation, emphasizing the need for accountable, transparent, and socially responsible AI systems to ensure inclusive and equitable societal outcomes. © The Author(s), under exclusive licence to Springer Nature Switzerland AG 2026.}, keywords = {Digital inclusion; Panel data analysis; Participatory urban innovation; Smart cities; Sustainable development; Urban governance}, } @article{hahnmsamhammerdtrettermdabrockp202684, title = {An actor-network-theory informed normative framework for the governance of AI-systems in medicine}, author = {Hahn M. and Samhammer D. and Tretter M. and Dabrock P.}, year = {2026}, journal = {Humanities and Social Sciences Communications}, volume = {13}, number = {1}, pages = {884}, doi = {10.1057/s41599-026-07957-8}, abstract = {The impact of Artificial Intelligence (AI) on clinical decision-making has become a focal point within a multitude of ethical, legal, and social discussions. Examining concrete use cases reveals that the introduction of an AI-system as a new actant in clinical practice changes the whole context with its relationships, roles, and routines. To better understand these changes, an interpretative framework can help to trace such transformations and uncover their ethical implications. In Science and Technology Studies (STS), a well-known approach is the application of Actor-Network-Theory (ANT) to concrete phenomena, including in the field of medicine. We aim to build on this established approach by developing a normative framework informed by ANT. To this end, we derive evaluation questions from Latour’s four meanings of mediation: translation to ask what goals should be pursued, composition to evaluate who should be involved, reversible blackboxing to specify how the different actants should be involved, and delegation to reflect on who should be responsible. These questions can thus inform the governance of AI-systems in medicine and contribute to their responsible implementation. © The Author(s) 2026.}, } @article{awanuahmedsalbishrinkaliszdszabszentgrtig202685, title = {Organisational responsible AI implementation and organisational foresight: The role of leadership control and managerial cognitive flexibility}, author = {Awan U. and Ahmed S. and Albishri N. and KALISZ D. and Szabó-Szentgróti G.}, year = {2026}, journal = {Technological Forecasting and Social Change}, volume = {227}, pages = {124623}, doi = {10.1016/j.techfore.2026.124623}, abstract = {Recent advances in artificial intelligence (AI) technologies have introduced new challenges related to privacy, transparency, and accountability, which have important implications for managerial decision-making. Although prior research has examined the adoption of AI in supply chain management, limited attention has been paid to how responsible AI interacts with managers' cognitive flexibility to shape organisational foresight. Drawing on Transaction Cost Economics, this study examines the combined effects of implementing responsible AI governance and leadership-enacted organisational control on the development of organisational foresight. This study contributes to organisational foresight research by explaining how the responsible implementation of responsible AI governance enhances organisations' ability to anticipate and interpret future developments through managerial cognitive flexibility. This research highlights responsible AI as a mechanism for balancing automation and human judgment, reinforcing the importance of human-centric governance in data-driven decision environments. Organisations should implement responsible AI alongside enabling leadership practices that encourage autonomy, experimentation, and reflective interpretation, as excessive control can undermine the cognitive flexibility necessary for developing organisational foresight. © 2026 The Authors}, keywords = {Decision making; Organisational foresight: Supply chain finance: Organisational control; Responsible artificial intelligence; Artificial intelligence; Behavioral research; Supply chain management; Supply chains; Technological forecasting; Artificial intelligence technologies; Chain management; Cognitive flexibility; Decisions makings; Managerial decision making; Organisational; Organizational controls; Organizational foresight: supply chain finance: organizational control; Responsible artificial intelligence; Supply chain finances; artificial intelligence; cognition; decision making; supply chain management; Decision making}, } @article{yigitcanlartliuksenadheerasmarasingherdavidacheongphcorchadoj202686, title = {Configurational pathways to smart city AI Adoption: Evidence from local governments in Australia, Hong Kong, Saudi Arabia, Spain, and the United States}, author = {Yigitcanlar T. and Liu K. and Senadheera S. and Marasinghe R. and David A. and Cheong P.H. and Corchado J.}, year = {2026}, journal = {Cities}, volume = {175}, pages = {107117}, doi = {10.1016/j.cities.2026.107117}, abstract = {Despite increasing policy attention and technological progress, AI adoption in smart city governance and local governments remains uneven. While previous studies have identified individual drivers of adoption, limited research has examined how multiple factors interact to enable or constrain implementation. Drawing on the technology-community-policy framework, this study employs fuzzy-set qualitative comparative analysis to investigate configurational pathways leading to AI-enabled smart city adoption across eleven local governments in five countries, Australia, Hong Kong, Saudi Arabia, Spain, and the United States. The findings reveal three different equifinal configurations leading to high AI adoption. First, the technology-driven pathway shows that robust smart city infrastructure and data capability can offset limited regulatory preparedness. Second, the balanced pathway integrates technological readiness, policy awareness, and organisational attention to community considerations to support adoption holistically. Third, the policy-driven pathway demonstrates that strong institutional mandates can compensate for weaker technical capacity. Across all pathways, perceived implementation constraints emerge as a core enabling condition, suggesting that recognition of challenges can stimulate proactive adoption strategies. The findings highlight substitutability between technological and policy dimensions, offering strategic flexibility for municipalities with differing resource endowments. This study advances configurational thinking in smart city and public sector innovation research and provides actionable insights for context-sensitive, resource-appropriate AI governance in local governments. © 2026 The Authors.}, keywords = {AI adoption; AI governance; Local government; Organisational configuration analysis; Public sector innovation; Smart city; Australia; China; Hong Kong; Saudi Arabia; Spain; United States; artificial intelligence; innovation; local government; smart city; technology adoption}, } @article{shenppengdchenyxuj202687, title = {The impact of AI social responsibility on service loyalty: a mixed-method approach}, author = {Shen P. and Peng D. and Chen Y. and Xu J.}, year = {2026}, journal = {Technology in Society}, volume = {87}, pages = {103375}, doi = {10.1016/j.techsoc.2026.103375}, abstract = {As artificial intelligence (AI) becomes increasingly integral to service industries, understanding how AI Social Responsibility (AISR) shapes customer relationships is critical. This paper aims to develop a dimensional framework of AISR and investigate the distinct mechanisms through which it influences service loyalty. A mixed-methods approach was employed, combining qualitative and quantitative techniques. First, in-depth interviews with 20 consumers who had used AI-enabled services were analyzed using Thematic Analysis, identifying two core dimensions: Technical AISR and Institutional AISR. Subsequently, two scenario-based experiments were conducted across different service contexts to test the proposed model. The results reveal a dual mediation mechanism: technical AISR primarily enhances service loyalty by boosting perceived AI competence, whereas institutional AISR primarily increases perceived AI empathy. Additionally, self-AI connection positively moderates these relationships, amplifying the effects of both AISR dimensions on their respective mediators. The paper concludes that AISR is a multidimensional construct that influences service loyalty through cognitive and affective pathways, with its effectiveness contingent on consumers' psychological bond with AI. These findings offer a novel theoretical lens for understanding AISR's role in service relationships and provide actionable guidance for firms seeking to strategically deploy responsible AI practices to cultivate lasting customer loyalty. © 2026 Elsevier Ltd.}, keywords = {AI social responsibility; Perceived AI competence; Perceived AI empathy; Self-AI connection; Service loyalty; Artificial intelligence; Service industry; Social aspects; Artificial intelligence social responsibility; Customer relationships; Mixed method; Perceived artificial intelligence competence; Perceived artificial intelligence empathy; Quantitative techniques; Self-artificial intelligence connection; Service industries; Service loyalty; Social responsibilities; artificial intelligence; corporate social responsibility; service sector; Economic and social effects}, } @article{husseinrzinkaramadanbhowardfmhightowermshahsbeaulieujonesbk202688, title = {Advancing healthcare AI governance through a comprehensive maturity model based on systematic review}, author = {Hussein R. and Zink A. and Ramadan B. and Howard F.M. and Hightower M. and Shah S. and Beaulieu-Jones B.K.}, year = {2026}, journal = {npj Digital Medicine}, volume = {9}, number = {1}, pages = {236}, doi = {10.1038/s41746-026-02418-7}, abstract = {Artificial Intelligence (AI) deployment in healthcare is accelerating, yet governance frameworks remain fragmented and often assume extensive resources. Through a systematic review of 35 frameworks for AI implementation in healthcare (published 2019–2024), we identified seven critical domains of healthcare AI governance. While existing frameworks provide valuable guidance, the resource requirements create barriers for smaller healthcare organizations. To address this gap, we organized key findings from the review to create the Healthcare AI Governance Readiness Assessment (HAIRA), a five-level maturity model that provides actionable governance pathways based on organizational resources. HAIRA spans from Level 1 (Initial/Ad Hoc) to Level 5 (Leading), with specific benchmarks across all seven governance domains. This tiered approach enables healthcare organizations to assess their current AI governance capabilities and establish appropriate advancement targets. Our framework addresses a critical need for adaptive governance strategies that ensure that AI implementation delivers tangible benefits to systems of varying resource levels. © The Author(s) 2026.}, keywords = {Artificial intelligence; Health care; Critical domain; Extensive resources; Healthcare organizations; Level-1; Maturity model; Model-based OPC; Organisational; Readiness assessment; Resource requirements; Systematic Review; article; artificial intelligence; benchmarking; health care; health care organization; human; maturity; systematic review; Reviews}, } @article{linzjinqlanj202689, title = {Empowering civic engagement in AI governance: A two-wave panel study on AI literacy and participatory governance of generative AI in China}, author = {Lin Z. and Jin Q. and Lan J.}, year = {2026}, journal = {Telecommunications Policy}, volume = {50}, number = {6}, pages = {103190}, doi = {10.1016/j.telpol.2026.103190}, abstract = {The rapid development of generative artificial intelligence (AI) has posed significant regulatory challenges. To examine civic engagement in AI governance, this study conceptualizes participatory governance of generative AI and examines its communicative and cognitive antecedents with a focus on AI literacy. Drawing on the Orientation-Stimulus-Reasoning-Orientation-Response (O-S-R-O-R) framework, we conducted a two-wave panel survey of generative AI users in China. Using cross-sectional structural models and an autoregressive cross-lagged panel model (ACLPM), we investigated directional and reciprocal influences over time. Results reveal that news consumption about generative AI increases AI policy knowledge, while interpersonal discussion builds AI knowledge. These two types of knowledge mutually reinforce each other and play distinct roles. AI knowledge acts as an empowering orientation, increasing participatory governance by enhancing efficacy. Conversely, AI policy knowledge acts as a contextual orientation, directly motivating participation and future news consumption but does not directly increase efficacy. Instead, participation itself increases efficacy over time.The results highlight a critical distinction between “empowered” and “informed” participation in China's state-led context. Furthermore, news consumption and participatory governance mutually reinforce one another. The findings demonstrate an upward, reinforcing spiral among media use, AI literacy, and participatory governance. This suggests that the O-S-R-O-R framework operates as a dynamic, reciprocal cycle rather than a static sequence. These findings further contribute to our understanding of the role of AI literacy in empowering and informing the public to engage with state-led AI governance. © 2026 Elsevier Ltd}, keywords = {AI governance; AI literacy; Efficacy; Interpersonal discussion; Knowledge; News consumption; Participatory governance; Artificial intelligence; Artificial intelligence governance; Artificial intelligence literacy; Civic engagement; Efficacy; Interpersonal discussion; Knowledge; News consumption; Panel studies; Participatory governance; Two waves; Public policy}, } @article{muralidharanasavulescuj202690, title = {All you need is…. justification: algorithmic justifiability trumps transparency}, author = {Muralidharan A. and Savulescu J.}, year = {2026}, journal = {Ethics and Information Technology}, volume = {28}, number = {2}, pages = {18}, doi = {10.1007/s10676-026-09892-3}, abstract = {Most ethical guidelines on AI tout algorithmic transparency, the openness of an algorithm’s inner workings to human scrutiny, as an important desideratum in algorithmic deployment. Algorithmic transparency has been touted as important for valuable goals like procedural fairness, AI trustworthiness, contestability and planning around AI decision-making. This paper argues that these goals are better served by a distinct desideratum, algorithmic justifiability, the ability of an algorithm to provide understanding about why the algorithm’s decision is correct. © The Author(s) 2026.}, keywords = {AI; Contestability; Justifiability; Procedural Fairness; Transparency; Trust; Artificial intelligence; Consensus algorithm; Decision making; Ethical technology; Algorithmics; Contestability; Decisions makings; Justifiability; Procedural fairness; Trust; Transparency}, } @article{yimmc202691, title = {The role of DEI-conscious PR in the AI-mediated communication era: Impact of employees’ diversity beliefs and inclusive climate on DEI-conscious AI usage intention}, author = {Yim M.C.}, year = {2026}, journal = {Public Relations Review}, volume = {52}, number = {3}, pages = {102703}, doi = {10.1016/j.pubrev.2026.102703}, abstract = {Generative AI (GenAI) chatbots allow direct user interaction, unlike traditional AI models. This direct engagement, combined with prompt repetition, risks reinforcing biases and normalizing certain communication styles that marginalize others. This study examines the impact of employees’ understanding of diversity, equity, and inclusion (DEI) values on their DEI-conscious intention to utilize AI chatbots, leveraging the social norm activation model as a framework. A survey (N = 280) revealed that when employees strongly believe in diversity and feel included within the organization, they are more likely to actively address AI bias and discrimination in their use of GenAI. The current study’s outcome makes a valuable contribution to expanding strategic employee communication literature at the intersection of AI and DEI. This study provides insights into how organizations can encompass DEI safeguards for employees’ chatbot usage at work. Once DEI values are fully embedded in corporate norms through employees’ diversity beliefs and an inclusive climate, employees are better positioned to embody these values when interacting with corporate stakeholders. © 2026 Elsevier Inc.}, keywords = {AI discrimination; DEI-conscious PR; Diversity belief; GenAI-mediated communication; Generative AI; Inclusive climate; Responsible AI; Strategic employee communication}, } @article{avramovinmarkoviaomitavokiszorninvujoviv202692, title = {PRIME–INSPECT: A Socio-Technical Framework for Trustworthy Intelligent Automation and Real-Time Decision-Making in Industry 4.0}, author = {Avramović N. and Marković A. and Čomić T. and Čavoški S. and Zornić N. and Vujović V.}, year = {2026}, journal = {Applied Sciences (Switzerland)}, volume = {16}, number = {10}, pages = {4825}, doi = {10.3390/app16104825}, abstract = {Intelligent automation is a core component of Industry 4.0, enabling artificial intelligence (AI) systems to support or execute operational and managerial decisions in real time. In high-risk industrial environments such as mining and metallurgy, real-time decision-making improves efficiency but also raises critical challenges related to trust, explainability, human oversight, and institutional accountability. This study proposes PRIME–INSPECT, a two-layer socio-technical framework designed to support trustworthy AI-driven real-time decision-making. The PRIME (predict, regulate, interpret, mitigate, execute) layer formalizes the operational decision flow, embedding control mechanisms, uncertainty quantification, and explainability into the automation pipeline. The INSPECT (integrity, navigability, supervisory control, policy maturity, ethical compliance, collaboration, trust calibration) layer defines the organizational and governance conditions required for safe deployment. The framework is conceptually developed through a structured literature synthesis and supported by exploratory empirical grounding through stakeholder perceptions from IT and top management participants, alongside an illustrative industrial use case intended to demonstrate conceptual applicability rather than engineering performance validation. The findings highlight the importance of aligning operational AI processes with institutional safeguards to support calibrated trust and responsible automation. The empirical component is intended to provide conceptual and organizational grounding of framework dimensions rather than quantitative validation of predictive performance. PRIME–INSPECT provides a structured architecture for designing and governing AI-enabled real-time decision systems in high-risk industrial contexts. © 2026 by the authors.}, keywords = {AI governance; explainable AI (XAI); high-risk industrial systems; human-in-the-loop; Industry 4.0; intelligent automation; real-time decision-making; trustworthy AI; Artificial intelligence; Automation; Decision making; Decision support systems; Industrial management; Real time systems; Uncertainty analysis; Verification; Artificial intelligence governance; Explainable artificial intelligence (XAI); High-risk industrial system; Human-in-the-loop; Industrial systems; Intelligent automation; Real time decision-making; Real-time decision making; Sociotechnical; Trustworthy artificial intelligence; article; artificial intelligence; automation; calibration; controlled study; decision making; ethical compliance; explainable artificial intelligence; human; metallurgy; mining; Compliance control}, } @article{bokharisaa202693, title = {Artificial intelligence, governance, and economic growth: Toward a new theory of digital developmentalism}, author = {Bokhari S.A.A.}, year = {2026}, journal = {Social Sciences and Humanities Open}, volume = {13}, pages = {102842}, doi = {10.1016/j.ssaho.2026.102842}, abstract = {This paper develops a new theoretical framework titled Digital Developmentalism, which explains how governments in the age of artificial intelligence (AI) and data-driven technologies are reshaping the mechanisms of economic growth and structural transformation. Integrating insights from developmental state theory, institutional economics, and general-purpose technology (GPT) theory, the study conceptualizes AI not merely as a technological innovation but as a transformative governance capability. The framework positions the state as a computational orchestrator, leveraging algorithmic coordination, data infrastructures, and public–private digital ecosystems to guide innovation and policy learning. Four key mechanisms underpin digital developmentalism: (1) algorithmic coordination, (2) data as developmental infrastructure, (3) AI-enhanced policy feedback, and (4) public–private digital synergy. Together, these mechanisms explain how AI governance capacity enables governments to manage complexity, enhance productivity, and foster inclusive development. The study also explores ethical and political challenges such as algorithmic opacity, surveillance, and inequality, emphasizing the need for transparent, accountable, and human-centered AI governance. By uniting governance theory and GPT theory into a single conceptual model, this paper contributes to the emerging discourse on how digital technologies redefine state capacity and the future of development. It concludes by proposing pathways for empirical testing and comparative research on AI-driven governance and economic performance across different national contexts. © 2026 The Author.}, keywords = {Algorithmic coordination; Artificial intelligence governance; Digital developmentalism; Economic growth; Institutional capacity}, } @article{kramerjjamilmwellsbjseidelmampejnairvgagenmcornpboviatayloryisrealmmcwilliamsa202694, title = {Building trust through transparency: Patient priorities, preferences, and guidance for effective AI disclosure}, author = {Kramer J. and Jamil M. and Wells B.J. and Seidel M. and Ampe J. and Nair V. and Gagen M. and Corn P. and Bovi A. and Taylor Y. and Isreal M. and McWilliams A.}, year = {2026}, journal = {Health Policy and Technology}, volume = {15}, number = {8}, pages = {101260}, doi = {10.1016/j.hlpt.2026.101260}, abstract = {Objectives: Health systems are rapidly expanding their use of artificial intelligence (AI) across the care continuum. However, best practices for explaining AI to patients to build trust and support positive experiences remain unclear. Methods: We conducted three waves of interviews with adult patients across a large health system, with each phase building upon prior findings. Phase 1 (N = 5) explored expectations for responsible AI use and communication preferences, which helped inform development of the system's AI evaluation framework. Phase 2 (N = 5) assessed patients’ informational needs following framework implementation and guided creation of draft patient-facing materials. Phase 3 (N = 8) used cognitive walkthroughs to evaluate these materials, which covered the system's approach to responsible AI and solution-specific applications. Transcripts were analyzed using thematic analysis and inductive and deductive coding. Results: Across phases, four themes emerged. Patients wanted (1) clear explanations of how AI supports clinical decision-making without replacing clinicians, (2) concrete examples of AI tasks, (3) clear rationales for adoption, and (4) transparent information about safety, privacy, and data protections. Phase 3 yielded actionable guidance for improving communications, including reducing technical or surveillance-oriented language, simplifying descriptions of AI functions, and centering patient safety and experience. Patients emphasized the need for straightforward explanations of data use and whether AI-supported care is optional, and the availability of plain-language, visually supported materials tailored to diverse literacy and language needs. Conclusions: Patients value clear and transparent communication about how AI supports care, particularly regarding safety, data use, and patient choice. Our findings offer practical guidance for developing patient-centered AI communication. Public interest summary: Artificial intelligence (AI) is becoming increasingly common in healthcare, but many patients are unsure and/or have concerns about what it means for their care. In this study, we interviewed patients from across a large health system to learn what information they want when AI tools are used by their care teams. Patients wanted simple explanations of what AI tools do, examples of how they are used, clear reasons for why they were adopted, and details about how AI supports their doctors and nurses but does not replace them. Trust also depended on patients knowing how their data are protected, who oversees the AI tools, and whether AI-supported care was optional. Patients highlighted the importance of straightforward materials that avoid jargon, offer visual aids, and are accessible to people of different backgrounds and abilities. These findings may help health systems explain AI in ways that improve patient understanding and trust. © 2026 Fellowship of Postgraduate Medicine}, keywords = {Ai; Artificial intelligence; Patients; Qualitative; Usability}, } @article{hamtinitm202695, title = {A Comprehensive Framework for Integrating Generative AI into Organizational Strategy}, author = {Hamtini T.M.}, year = {2026}, journal = {TEM Journal}, volume = {15}, number = {2}, pages = {1086-1095}, doi = {10.18421/TEM152-13}, abstract = {Generative AI enables autonomous content creation and fosters innovation, yet many organizations encounter fragmented governance, limited infrastructure, and cultural resistance. This research investigates how organizations can strategically integrate Generative AI to achieve lasting transformation. Drawing on literature in AI governance and organizational maturity models, a three-level framework (Foundation, Optimization, and Transformation) was developed. The study adopted a mixed-methods approach, combining qualitative interviews and quantitative metrics to assess outcomes. Using a case-study approach within an education technology organization, implementation of the framework resulted in measurable gains in customer retention. The company’s retention rate increased from 62% to 78.5% over six months, a 16.5% gain which contributed directly to increased recurring revenue. Improved engagement, platform satisfaction, and personalized GenAI-powered tutoring features underpinned these outcomes. These findings support the hypothesis that a phased maturity model enhances both strategic alignment and cultural readiness. Further validation across diverse sectors is recommended to ensure scalability and adaptability in dynamic AI environments. © 2026 Thair M. Hamtini}, keywords = {digital transformation; Edtech; Generative AI; Maturity Model; organizational strategy}, } @article{lipskiynflowerdaysv202696, title = {Healthcare AI as Critical Digital Health Infrastructure: A Public Health Preparedness Framework for Systemic Risk}, author = {Lipskiy N. and Flowerday S.V.}, year = {2026}, journal = {Future Internet}, volume = {18}, number = {5}, pages = {232}, doi = {10.3390/fi18050232}, abstract = {Healthcare artificial intelligence (AI) is moving from the laboratory into the infrastructure of care. As these systems become embedded in imaging, electronic health records, triage, and clinical decision support, their failures can affect not only individual encounters but also institutions and patient populations. Yet governance still centers on model development, local validation, and one-time compliance, with limited attention to cross-site failure after deployment. This article examines how public health preparedness can help close that gap. It presents a conceptual analysis grounded in two cases: a pneumonia-screening convolutional neural network that learned institutional confounders rather than portable clinical signals, and a widely deployed sepsis prediction model whose external performance and alert burden fell short of developer claims. Together, these cases reveal five governance features of systemic healthcare AI risk: population-level exposure, cascade effects across shared infrastructures, unequal vulnerability, delayed recognition, and coordination needs beyond any single institution. In response, we propose a tripartite framework combining stronger pre-deployment assurance, post-deployment surveillance with escalation thresholds, and tertiary response through investigation, rollback, remediation, and cross-site learning. The argument is not that AI failures are epidemics, but that high-impact clinical AI systems now function as critical digital health infrastructure requiring preparedness alongside lifecycle oversight. © 2026 by the authors.}, keywords = {algorithm vigilance; collaborative surveillance; cyber–physical systems; health infrastructure; healthcare AI governance; public health; public health intelligence; sociotechnical safety; systemic healthcare AI risk; Artificial intelligence; Critical infrastructures; Diagnosis; Electronic health record; Health care; Health risks; Life cycle; Public risks; Algorithm vigilance; Collaborative surveillance; Cybe-physical systems; Cyber-physical systems; Health infrastructure; Healthcare artificial intelligence governance; Public health intelligence; Sociotechnical; Sociotechnical safety; Systemic healthcare artificial intelligence risk; Public health}, } @article{kramerjnairvwellsbjgagenmisrealmcornpnguyenhpallinimboviataylorychoushhetheringtontmcwilliamsa202697, title = {Leveraging stakeholder engagement to develop and evaluate a responsible artificial intelligence framework at a large, multi-state health system}, author = {Kramer J. and Nair V. and Wells B.J. and Gagen M. and Isreal M. and Corn P. and Nguyen H. and Pallini M. and Bovi A. and Taylor Y. and Chou S.-H. and Hetherington T. and McWilliams A.}, year = {2026}, journal = {Health Policy and Technology}, volume = {15}, number = {7}, pages = {101243}, doi = {10.1016/j.hlpt.2026.101243}, abstract = {Objectives: Artificial intelligence (AI) offers health systems opportunities to enhance care delivery, improve efficiency, and expand patient access. However, rapid innovation introduces new risks requiring careful oversight. This study examines how diverse stakeholders shaped the design and early evaluation of the Framework for the Appropriate Implementation and Review of AI (FAIR-AI), a system-wide AI governance framework implemented within a large, multi-state health system. Methods: We conducted two rounds of semi-structured interviews – before FAIR-AI development and shortly after FAIR-AI was approved – with executive leaders (N = 5), risk/compliance/legal leaders (N = 11), and data developers (N = 8) to identify initial design needs and evaluate the approved framework. Pre-development interviews also included patients (N = 5) and clinicians (N = 5) to capture AI end-user expectations. Data were analyzed using thematic analysis and inductive and deductive coding methodologies. Results: Pre-development interviews highlighted three central priorities: balancing risk tolerance with potential benefits, ensuring direct human oversight, and streamlining review for low-risk solutions. Patients and clinicians emphasized the need for clinician control over care decisions, with AI serving as supplemental support. Post-approval interviews identified seven elements critical to success: (1) transparent and consistent reviews; (2) timely evaluations; (3) ongoing solution monitoring; (4) iterative framework refinement; (5) alignment with institutional priorities and regulatory standards; (6) multi-modal teammate education; and (7) diverse patient dissemination efforts. Conclusions: Our findings highlight the importance of AI governance frameworks integrating both pre-deployment risk assessment and post-implementation solution monitoring, while remaining adaptable through feedback loops and in response to changing regulatory and technological contexts. This stakeholder-informed approach provides practical guidance for responsible AI at enterprise scale. © 2026 Fellowship of Postgraduate Medicine}, keywords = {Ai framework; Artificial intelligence; Health Services Research; Health system administrators; Qualitative; administrative personnel; article; artificial intelligence; clinician; diagnosis; human; interview; postgraduate student; responsible artificial intelligence; risk assessment; semi structured interview; stakeholder engagement; thematic analysis}, } @article{sotolanirfreiresfronchettifsantosrdsspinolar202698, title = {Exposing hidden bias: A study of fairness debt in gray literature}, author = {Sotolani R. and Freire S. and Fronchetti F. and Santos R.D.S. and Spinola R.}, year = {2026}, journal = {Journal of Systems and Software}, volume = {238}, pages = {112866}, doi = {10.1016/j.jss.2026.112866}, abstract = {Context: Bias in AI and software systems has become a major concern due to its potential to reinforce discrimination, leading to the notion of software fairness debt. Objective: This study investigates how fairness debt is discussed in gray literature, with the goal of identifying concrete examples, underlying causes, observable effects, and proposed solutions. Method: We applied a query-based retrieval strategy to collect 79 practitioner-oriented articles and conducted content analysis guided by an established fairness debt conceptual model. Results: Publicly documented fairness problems are overwhelmingly concentrated in AI-based systems. We identified 19 instances of fairness debt (e.g., racism, sexism, ageism), 26 causes (e.g., biased training data, lack of team diversity, historical inequalities, organizational pressures), 13 effects (e.g., reinforcement of stereotypes, systemic exclusion, erosion of trust, reputational harm), and 18 proposed solutions (e.g., bias detection methods, fairness-aware design, independent audits, stakeholder participation, ethical guidelines). These findings show that fairness debt is a multifaceted phenomenon with intertwined technical, organizational, and societal dimensions. Conclusion: This study extends the fairness debt framework with evidence from non-academic sources, revealing how fairness issues arise and are addressed in practice. For researchers, it highlights the need to study fairness debt as a multidimensional phenomenon and to advance methods for its assessment and mitigation. For practitioners, it underscores the importance of embedding fairness into development lifecycles, adopting inclusive design, and strengthening organizational accountability. Together, these insights bridge theory and practice, offering a foundation for more effective strategies to reduce fairness debt in software systems. and overviews. CCS Concepts: • General and reference → Surveys © 2026 Elsevier Inc.}, keywords = {Fairness debt; Gray literature; Software fairness; Computer software; Ethical technology; AI systems; Conceptual model; Content analysis; Fairness debt; Grey literature; Organisational; Retrieval strategies; Software fairness; Software-systems; Underlying cause; Life cycle}, } @article{agrawallmulgundpdasouzarobhayaksinghr202699, title = {AI Regulation in U.S. States: Lessons Learned and Key Takeaways}, author = {Agrawal L. and Mulgund P. and Dasouza R.O. and Bhaya K. and Singh R.}, year = {2026}, journal = {Communications of the ACM}, volume = {69}, number = {6}, pages = {68-77}, doi = {10.1145/3778178}, abstract = {U.S. states have rapidly attempted to fill the regulatory void created by the lack of comprehensive federal legislation, leading to a dramatic surge in state-level artificial intelligence (AI)-related bills. Between 2019 and 2024, lawmakers introduced nearly 803 AI bills; yet only 127 have been enacted, and 17 have been adopted as resolutions. However, this state-driven legislative activity has resulted in an unprecedented and fragmented regulatory landscape. Our analysis of all AI-focused bills across U.S. states identifies the prominent legislative themes, including government use of AI, regulation of AI in the private sector, responsible AI practices, bias mitigation, workforce development, child protection, consumer protection, and the creation of advisory bodies or studies, highlighting widespread concerns about transparency, accountability, and societal risks. Additionally, we observe striking variation across states and political alignments. A handful of states, notably New York, New Jersey, Illinois, and Massachusetts, account for a disproportionately large share of pending bills, while pioneering states such as Colorado, Utah, and California have enacted substantive laws addressing distinct facets of AI governance. This decentralized approach validates the notion of states as "laboratories" of regulation, but also underscores the risks associated with fragmented rules. Lastly, we compare these diverse state-level initiatives with the European Union's uniform AI Act and discuss their implications for shaping future federal AI policies. © 2026 Copyright held by the owner/author(s).}, keywords = {Artificial intelligence; Personnel; Risk assessment; Advisory bodies; Child protections; Federal legislations; Illinois; Legislative activities; New Jersey; New York; Private sectors; Societal risks; Workforce development; Consumer protection}, } @article{lilmatdesamablitliw2026100, title = {Overview of the Application of Generative Artificial Intelligence in Film Production: Algorithms, Tools, and Future Trends}, author = {Li L. and Mat Desa M.A.B. and Li T. and Li W.}, year = {2026}, journal = {Studies in Media and Communication}, volume = {14}, number = {2}, pages = {22-39}, doi = {10.11114/smc.v14i2.8095}, abstract = {The rapid evolution of generative artificial intelligence (GenAI) is transforming the film industry. This article reviews key GenAI algorithms, including GANs, VAEs, diffusion models, and transformer-based architectures, and explores their application across various stages of film production, from scriptwriting to post-production. Through case studies such as The Frost and Netflix's AI-assisted projects, the study illustrates GenAI's creative potential and workflow innovations. It also addresses critical ethical and legal concerns, including authorship disputes, deepfakes, and algorithmic bias. Finally, the paper outlines future directions, such as multimodal model integration and AI-human co-creation, advocating for a responsible and human-centered implementation of these technologies. Copyright for this article is retained by the author(s), with first publication rights granted to the journal. This is an open-access article distributed under the terms and conditions of the Creative Commons Attribution license (http://creativecommons.org/licenses/by/4.0/).}, keywords = {AI in cinema; creative automation; diffusion models; ethical AI; film production; generative AI; multimodal models}, } @article{wangwliz2026101, title = {The Evolution of Artificial Intelligence in Marketing: A Bibliometric Analysis of Three Decades (1992–2025)}, author = {Wang W. and Li Z.}, year = {2026}, journal = {Informatics}, volume = {13}, number = {5}, pages = {67}, doi = {10.3390/informatics13050067}, abstract = {Over the past three decades, artificial intelligence (AI) has substantially reshaped marketing research and practice, yet the discipline has not established a systematic understanding of its evolutionary trajectory and intellectual structure. A bibliometric analysis of 1923 Scopus publications (1992–2025) was conducted using CiteSpace to explore collaboration patterns, conceptual development, and thematic organization. It identified six evolutionary stages with accelerating innovation cycles, starting with neural networks (1992–2000) and ending with generative AI (2024–2025), with research attention per stage compressing from approximately 9 years to just 2 years. The analysis of the collaboration network shows that the key contributors are India, China, the USA, and the UK. Co-citation analysis indicates that there are three thematic dimensions with seven clusters, namely: (i) AI technological foundations and capabilities, (ii) AI marketing applications and transformation, and (iii) responsible AI governance and ethics. It suggests a Three-Force Evolutionary Framework, which combines technology-push, market-pull, and governance-moderator forces to describe the dynamics of the field. This framework shows that the Regulatory Awakening of 2018 (e.g., GDPR and the Cambridge Analytica incident) guided, not limited, innovation, and highlighted the critical personalization–privacy paradox on which modern developments are based. It identifies three priority research directions: generative AI in creative marketing, consumer trust in the personalization–privacy paradox, and organizational adaptation to fast innovation cycles. This study provides scholars with a comprehensive knowledge map, practitioners with strategic imperatives for responsible AI adoption, and policymakers with evidence that well-designed regulation accelerates innovation by balancing commercial value with societal concerns. © 2026 by the authors.}, keywords = {AI; artificial intelligence; bibliometric analysis; CiteSpace; generative AI; marketing}, } @article{yunismjamalidumarm2026102, title = {The illusion of AI mastery: “Skill Inflation” in the age of accessible AI tools}, author = {Yunis M. and Jamali D. and Umar M.}, year = {2026}, journal = {Journal of Innovation and Knowledge}, volume = {18}, pages = {101084}, doi = {10.1016/j.jik.2026.101084}, abstract = {As artificial intelligence (AI) becomes increasingly embedded in organizational processes, hiring systems, and digital transformation agendas, a critical but underexamined challenge has emerged: AI skill inflation, or the overstatement of AI-related competence relative to demonstrable capability. This phenomenon is distinct from simple AI literacy gaps or credential accumulation because it reflects a broader claim-capability misalignment shaped by psychological, social, and institutional forces. This paper develops an integrated conceptual framework to explain how AI skill inflation emerges, diffuses, and becomes normalized in contemporary workplaces. Drawing on signaling theory, social cognitive theory, institutional theory, impression management, and the Dunning-Kruger account of overconfidence and miscalibration, we argue that low-barrier credentials, accessible AI tools, media hype, and organizational transformation pressures create favorable conditions for inflated competence claims. These dynamics are further amplified by self-enhancement, peer reinforcement, and digital self-presentation, while AI literacy and regulatory environments condition their strength and consequences. The framework advances five propositions and highlights the organizational, ethical, and governance risks of AI skill inflation, including hiring mismatches, project failure, compliance exposure, and erosion of trust in AI initiatives. By introducing AI skill inflation as a distinct organizational construct, this study contributes to the literature on AI governance, responsible innovation, and capability development, while providing a foundation for future empirical research and practical interventions in hiring, training, and oversight. © 2026 The Author(s).}, keywords = {AI literacy; AI skill inflation; Artificial intelligence; Digital transformation; Impression management; Signaling theory}, } @article{moosavihaghighim2026103, title = {Modeling AI-driven inequality and adaptive governance: A system dynamics approach to U.S. Socioeconomic futures}, author = {Moosavihaghighi M.}, year = {2026}, journal = {Sustainable Futures}, volume = {11}, pages = {101746}, doi = {10.1016/j.sftr.2026.101746}, abstract = {Artificial Intelligence is reshaping socioeconomic systems by enhancing productivity while intensifying concerns about inequality, unemployment, and policy responsiveness. This study employs a System Dynamics model to simulate the U.S. socioeconomic landscape from 2000 to 2035, focusing on the interdependencies between AI investment, income distribution, and adaptive policy design. Given data constraints, AI investment is modeled as a uniform labor market driver, with international competition introduced via the DeepSeek stress test. The model integrates political feedback loops linking wealth concentration to reform inertia. Three policy scenarios are evaluated: (1) baseline U.S. AI adoption, (2) competitive pressure from a low-cost foreign platform under varying regulations, and (3) adaptive reforms coupling AI taxation and redistribution to real-time inequality and unemployment metrics. Results reveal that while AI-driven productivity may reduce unemployment and cost of production initially, it exacerbates inequality without responsive governance. Adaptive mechanisms, such as dynamic reskilling and AI-linked fiscal tools, outperform static interventions in promoting equity and competitiveness. However, entrenched political influence constrains reform unless public dissatisfaction crosses critical thresholds. These findings highlight the urgent need for anticipatory, adaptive policy frameworks that align technological innovation with inclusive and sustainable socioeconomic outcomes. © 2026 The Author. Published by Elsevier Ltd. This is an open access article under the CC BY license. http://creativecommons.org/licenses/by/4.0/}, keywords = {Adaptive Governance; AI Governance; Artificial Intelligence; Income Inequality; Labor Market; System Dynamics}, } @article{adibkletchfordndunninghesalamantoliasydebarrosjazzopardimuscatnklugehhpnovilloortizd2026104, title = {Governance of artificial intelligence for health systems, WHO European Region}, author = {Adib K. and Letchford N. and Dunning H.E. and Salama N. and Tolias Y. and De Barros J. and Azzopardi-Muscat N. and Kluge H.H.P. and Novillo-Ortiz D.}, year = {2026}, journal = {Bulletin of the World Health Organization}, volume = {104}, number = {6}, pages = {374-385}, doi = {10.2471/BLT.25.294978}, abstract = {الغرض: تقديم نظرة عامة عن حالة حوكمة الذكاء الاصطناعي (AI) في مجال الصحة في المنطقة الأوروبية لمنظمة الصحة العالمية (WHO). الطريقة: أعد المكتب الإقليمي لمنظمة الصحة العالمية لأوروبا مسحًا يشمل عدة قطاعات حول حوكمة الذكاء الاصطناعي وتطبيقه. وأتيح المسح لجميع الدول الأعضاء البالغ عددها 53 دولة في الفترة ما بين يونيو/حزيران 2024، ومارس/آذار 2025. تم تحليل الردود على المستويين الإقليمي وتحت الإقليمي. النتائج: من بين الدول الأعضاء الـ 50 التي شاركت في المسح، كان لدى %8 (4/50) منها استراتيجية خاصة بالذكاء الاصطناعي في مجال الصحة، وتسعى %14 (7) منها لوضع استراتيجية لها. على الصعيد الإقليمي، قامت 33 دولة عضو بتنفيذ استراتيجيات للذكاء الاصطناعي تشمل عدة قطاعات، وتقوم 8 دول (%16) لوضع استراتيجية. ومن بين هذه الدول الأعضاء، كلف ما يقرب من نصفها وكالة حكومية، قائمة واحدة أو أكثر، بمسؤوليات التنفيذ والرقابة. وأفاد ما يقرب من نصف الدول الأعضاء (23) بإجراء تقييمات مستمرة للقوانين والسياسات المتعلقة بأنظمة الذكاء الاصطناعي، ووضع خُمسها (10) قوانين جديدة خاصة بالذكاء الاصطناعي في مجال الصحة. ولم تصدر سوى 14 دولة عضو مبادئ توجيهية لمواجهة الآثار الأخلاقية لاستخدام الذكاء الاصطناعي في مجال الصحة أو عبر القطاعات. ووضعت أقل من %10 (4) من الدول الأعضاء معايير للمسؤولية فيما يتعلق بالذكاء الاصطناعي أو إرشادات بشأن تطبيق معايير المسؤولية القائمة. بوجه عام، اعتمدت 33 دولة عضو (%66) استراتيجيات للبيانات الصحية، ولدى 33 دولة عضو (%66) مراكز للبيانات الصحية، ولدى %68 منها هيئات معنية بالبيانات الصحية تدعم الوصول إلى البيانات ومراقبتها. الاستنتاج: في المنطقة الأوروبية لمنظمة الصحة العالمية، لا تزال حوكمة الذكاء الاصطناعي في مجال الرعاية الصحية في مرحلة متأخرة. وهناك حاجة إلى آليات قانونية وسياسية قابلة للتكيف من أجل الاستجابة بفعالية للتحديات المعقدة والمتغيرة التي ينطوي عليها دمج الذكاء الاصطناعي في النظم الصحية.; 目的: 旨在概述世卫组织 (WHO) 欧洲区域在卫生领域的人工智能 (AI) 治理现状。. 方法: 世卫组织欧洲区域办事处就 AI 治理和采用问题开展了一项横断面研究。在 2024 年 6 月至 2025 年 3 月期间,该项研究向 53 个会员国全面开放。我们在区域和次区域层面对研究结果进行了分析。. 结果: 在参与该项研究的 50 个会员国中,有 8% (4/50) 已实施了针对卫生领域的 AI 战略,另有 14%(7 个)正处于战略制定阶段。在区域层面,有 33 个会员国已实施了跨部门 AI 战略,另有 8 个(占 16%)正处于战略制定阶段。在这些会员国中,已将实施职责和监督职责分别分配给一个或多个既有政府机构的会员国几乎各占一半。有近一半会员国(23 个)称其对 AI 系统相关法律和政策进行了持续评估,另有五分之一会员国(10 个)已制定了新的针对卫生领域的 AI 法律。仅 14 个会员国发布了旨在应对在卫生部门或跨部门使用 AI 所涉道德问题的指南。不到 10%(4 个)的会员国已制定了针对 AI 的责任标准或发布了有关如何应用现有责任标准的指南。总体而言,有 33 个会员国(占 66%)推行了卫生数据相关战略;有 33 个会员国(占 66%)建立了卫生数据中心;以及有 68% 的会员国成立了提供数据获取与控制服务的卫生数据管理机构。. 结论: 世卫组织欧洲区域在卫生保健领域的 AI 治理框架尚不完善。需要构建自适应法律和政策机制,以有效应对在将 AI 集成至卫生系统过程中面临的不断演变的复杂挑战。.; Objective: To provide an overview of the status of artificial intelligence (AI) governance for health in the World Health Organization (WHO) European Region. Methods: The WHO Regional Office for Europe developed a cross-sectional survey on AI governance and adoption. The survey was available to all 53 Member States between June 2024 and March 2025. Responses were analysed at the regional and subregional level. Findings: Of the 50 Member States responding to the survey, 8% (4/50) have a health-specific AI strategy and 14% (7) are developing one. Regionally, 33 Member States have implemented cross-sectoral AI strategies and 8 (16%) are developing them. Of these Member States, almost half each have assigned implementation and oversight responsibilities to existing single or multiple government agencies. Nearly half the Member States (23) reported ongoing assessments of laws and policies on AI systems and a fifth (10) have developed new health-specific AI laws. Only 14 Member States have issued guidelines to address the ethical implications of using AI in health or across sectors. Less than 10% (4) of Member States have developed liability standards for AI or guidance on the application of existing liability standards. Overall, 33 (66%) Member States have adopted health data strategies, 33 (66%) have health data hubs and (68%) have health data authorities supporting data access and control. Conclusion: In the WHO European Region, governance of AI in health care is underdeveloped. Adaptive legal and policy mechanisms are needed to respond effectively to the complex and evolving challenges of AI integration in health systems. (c) 2026 The authors; licensee World Health Organization.; Objectif: Offrir un aperçu de l’état de la gouvernance de l’intelligence artificielle (IA) dans le domaine de la santé au sein de la région Europe de l’Organisation mondiale de la Santé (OMS). Méthodes: Le Bureau régional de l’OMS pour l’Europe a élaboré une enquête transversale sur la gouvernance et l’adoption de l’IA. Cette enquête a été mise à la disposition des 53 États membres entre juin 2024 et mars 2025. Les réponses ont été analysées aux niveaux régional et sous-régional. Résultats: Sur les 50 États membres ayant répondu à l’enquête, 8% (4/50) disposent d’une stratégie d’IA spécifique au domaine de la santé et 14% (7) en sont en train d’élaborer une. À l’échelle régionale, 33 États membres ont mis en œuvre des stratégies intersectorielles en matière d’IA et 8 (16%) sont en train de les élaborer. Parmi ces États membres, près de la moitié ont confié les responsabilités de mise en œuvre et de surveillance à une ou plusieurs agences gouvernementales existantes. Près de la moitié des États membres (23) ont indiqué mener actuellement des évaluations des lois et politiques relatives aux systèmes d’IA, et un cinquième (10) ont élaboré de nouvelles lois sur l’IA qui sont spécifiques au domaine de la santé. Seuls 14 États membres ont publié des lignes directrices visant à traiter les implications éthiques de l’utilisation de l’IA dans le domaine de la santé ou dans d’autres secteurs. Moins de 10% (4) des États membres ont mis au point des normes de responsabilité pour l’IA ou des orientations sur l’application de normes de responsabilité existantes. Dans l’ensemble, 33 États membres (66%) ont adopté des stratégies en matière de données de santé, 33 (66%) disposent de plateformes de données de santé et 68% ont mis en place des autorités chargées des données de santé qui facilitent l’accès aux données et leur régulation. Conclusion: Dans la région Europe de l’OMS, la gouvernance de l’IA dans les soins de santé est insuffisamment développée. Des mécanismes juridiques et politiques adaptatifs sont nécessaires pour répondre efficacement aux défis complexes et en constante évolution liés à l’intégration de l’IA dans les systèmes de santé.; Цель: Дать обзор состояния управления применением искусственного интеллекта (ИИ) для целей здравоохранения в Европейском регионе Всемирной организации здравоохранения (ВОЗ). Методы: Европейское региональное бюро ВОЗ разработало опрос в рамках поперечного исследования по вопросам управления в сфере ИИ и его внедрения. Опрос проводился среди всех 53 государств-членов в период с июня 2024 года по март 2025 года. Ответы были проанализированы на региональном и субрегиональном уровнях. Результаты: Из 50 государств-членов, принявших участие в опросе, 8% (4 из 50) имеют профильную стратегию ИИ в сфере здравоохранения и 14% (7) разрабатывают ее. На региональном уровне 33 государства-члена внедрили межсекторальные стратегии ИИ и 8 государств (16%) разрабатывают их. Почти в половине этих государств-членов обязанности по внедрению и надзору возложены на одно существующее ведомство, а в другой половине – на несколько. Почти половина государств-членов (23) сообщили о проведении оценки действующего законодательства и мер политики в отношении систем ИИ, и пятая часть (10) разработала новое законодательство относительно ИИ специально для сферы здравоохранения. Только 14 государств-членов разработали рекомендации по решению этических аспектов использования ИИ в здравоохранении или различных секторах. Менее 10% (4) государств-членов разработали стандарты юридической ответственности в сфере ИИ или рекомендации по применению имеющихся стандартов юридической ответственности. В целом 33 государства-члена (66%) внедрили стратегии относительно данных в здравоохранении, 33 (66%) имеют центры хранения и обработки данных по здравоохранению, в 68% созданы профильные органы, обеспечивающие поддержку доступа к медицинским данным и контроль над ними. Вывод: В Европейском регионе ВОЗ система управления применением ИИ в здравоохранении развита недостаточно. Необходимы адаптивные правовые и политические механизмы для эффективного реагирования на сложные и постоянно меняющиеся задачи, связанные с интеграцией ИИ в системы здравоохранения.; Objetivo: Proporcionar una visión general del estado de la gobernanza de la inteligencia artificial (IA) para la salud en la Región Europea de la Organización Mundial de la Salud (OMS). Métodos: La Oficina Regional de la OMS para Europa desarrolló una encuesta transversal sobre la gobernanza y la adopción de la IA. La encuesta estuvo disponible para los 53 Estados Miembros entre junio de 2024 y marzo de 2025. Las respuestas se analizaron a nivel regional y subregional. Resultados: De los 50 Estados Miembros que respondieron a la encuesta, el 8% (4/50) cuenta con una estrategia de IA específica para la salud y el 14% (7) se encuentra desarrollando una. A nivel regional, 33 Estados Miembros han implementado estrategias de IA intersectoriales y 8 (16%) se encuentran en proceso de desarrollarlas. De estos Estados Miembros, casi la mitad han asignado responsabilidades de implementación y supervisión a una o varias agencias gubernamentales existentes. Casi la mitad de los Estados Miembros (23) informaron de evaluaciones en curso de leyes y políticas sobre sistemas de IA, y una quinta parte (10) ha desarrollado nuevas leyes de IA específicas para la salud. Solo 14 Estados Miembros han emitido directrices para abordar las implicaciones éticas del uso de la IA en la salud o en distintos sectores. Menos del 10% (4) de los Estados Miembros ha desarrollado normas de responsabilidad para la IA o directrices sobre la aplicación de normas de responsabilidad existentes. En conjunto, 33 (66%) Estados Miembros han adoptado estrategias de datos sanitarios, 33 (66%) disponen de repositorios de datos sanitarios y el 68% cuenta con autoridades de datos sanitarios que apoyan el acceso y el control de los datos. Conclusión: En la Región Europea de la OMS, la gobernanza de la IA en la asistencia sanitaria está poco desarrollada. Se requieren mecanismos jurídicos y de política adaptativos para responder eficazmente a los desafíos complejos y cambiantes de la integración de la IA en los sistemas sanitarios.}, keywords = {Artificial Intelligence; Cross-Sectional Studies; Delivery of Health Care; Europe; Humans; World Health Organization; artificial intelligence; cross-sectional study; Europe; health care delivery; human; organization and management; World Health Organization}, } @article{yadavssamadhiyaakumarapandeykkluthraseljaouharia2026105, title = {Generative AI-driven sustainability in supply chains: A micro foundation of dynamic capability towards a socially responsible supply chain to achieve greater societal change}, author = {Yadav S. and Samadhiya A. and Kumar A. and Pandey K.K. and Luthra S. and El jaouhari A.}, year = {2026}, journal = {Technological Forecasting and Social Change}, volume = {229}, pages = {124726}, doi = {10.1016/j.techfore.2026.124726}, abstract = {The application of Gen AI (Generative AI) across multiple sectors like manufacturing and service domains, shows transformative effects to improve socially responsible decision-making and collaborative efforts. Yet it remains insufficiently investigated in the context of a socially responsible supply chain (SRSC) towards sustainable supply chain management (SSCM) in a wider context. Gen AI enables faster reporting and adaptive responses to enhance decision-making, which together improve supply chain flexibility while promoting social responsibility. Although previous research recognizes Gen AI's contribution to social functionality within a supply chain, it does not provide a full theoretical structure for analyzing how Gen AI solutions develop and function in SSCM. Prior research stresses the importance of making people and communities central elements in SSCM from the outset. To address this gap, this research conducts a rigorous qualitative study by analyzing 82 exemplary SSCM cases from manufacturing and service sectors through content analysis. The research explores how organizations can leverage dynamic capability theory (DCT) to adopt and integrate Gen AI systems. The findings demonstrate the stakeholder role in SSCM: 1) NGOs and universities provide essential knowledge and skills together with resources which support sustainable practices; 2) active collaboration with external stakeholders creates competitive benefits while promoting wider implementation of sustainability efforts through imitation. This research delivers a conceptual framework, showing how dynamic supply chain capabilities enabled by Gen AI affect stakeholder alignment towards sustainability goals while mobilizing stakeholders towards SSCM practices; this creates positive effects for wider communities in dynamically evolving Gen AI based SC systems. Our study utilizes micro-foundations of dynamic capabilities to deliver actionable recommendations for managers and outlines future research paths for expanding sustainability practices across multiple dimensions using Gen AI. This study provides helpful insights for professionals, researchers, and leaders to achieve Sustainable Development Goals (SDGs). © 2026 Elsevier Inc.}, keywords = {Capability reconfiguration; Digital sustainability transition; Responsible AI adoption; Social change; Socio-technical innovation; Supply chain resilience; Behavioral research; Competition; Decision making; Economic and social effects; Social aspects; Supply chain management; Supply chains; Sustainable development; Sustainable development goals; Capability reconfiguration; Digital sustainability transition; Responsible AI adoption; Social changes; Socio-technical innovation; Sociotechnical; Supply chain resiliences; Sustainability transition; Sustainable supply chains; Technical innovation; artificial intelligence; competition (economics); decision making; innovation; manufacturing; service sector; social change; supply chain management; sustainability; technology adoption; theoretical study; Industrial research}, } @article{lowmptr2026106, title = {Artificial intelligence and performance outcomes: a two-wave study of employees and organizations}, author = {Low M.P. and T R.}, year = {2026}, journal = {Academia Revista Latinoamericana de Administracion}, pages = {1-24}, doi = {10.1108/ARLA-02-2025-0031}, abstract = {Purpose – This study examines the impact of AI on employee and organizational performance, focusing on four employee factors (AI awareness, evaluation, usage, ethics) and four organizational factors (business spanning, infrastructure, proactive stance, ethical considerations) that drive overall performance in today's business environment. Design/methodology/approach – The research uses a two-wave study design, collecting data from businesses using purposive sampling technique. Study 1 focuses on AI at the employee level, with 222 matched data points to examine how AI literacy affects employee performance. Study 2 examines organizational AI competency, with 179 matched data points assessing impact on organizational performance. Data gathered was analyzed using PLS-SEM for explanatory and predictive purposes. Findings – At the employee level, the results indicate that AI literacy enhances employee performance, with IPMA and NCA highlighting AI usage, awareness, and evaluation as key drivers. At the organizational level, proactive stance, infrastructure, and business spanning are essential for performance, while AI ethics matters primarily at peak performance. The two-wave study is supported by robust analyses, including endogeneity tests, IPMA, and NCA. Research limitations/implications – This study is limited by its purposive sampling technique, the absence of comparison groups and control variables. Originality/value – This two-wave study develops a multi-level framework linking employee and organizational AI capabilities to improve organizational performance. © Emerald Publishing Limited}, keywords = {AI competency; AI literacy; Alfabetización en IA; Competencia en IA; Desempeño de los empleados; Desempeño organizacional; Employee performance; Estudio de dos fases; Organizational performance; Two-wave study}, } @article{chenggzhangh2026107, title = {The impact of china’s artificial intelligence pilot policies on enterprise supply chain resilience}, author = {Cheng G. and Zhang H.}, year = {2026}, journal = {Scientific Reports}, volume = {16}, number = {1}, pages = {5382}, doi = {10.1038/s41598-025-32003-z}, abstract = {As the “Artificial Intelligence Plus” strategic initiative continues to deepen, the circumstances under which and how artificial intelligence (AI) can enhance the resilience of corporate supply chains are rapidly drawing academic attention. This paper, utilizing data from 4,144 A-share listed companies in China and relevant data from prefecture-level cities spanning from 2016 to 2023 as samples, employs the double machine learning (DML) method with random forest regression as the DML learner to investigate the relationship mechanism between government-level AI policies and corporate supply chain resilience. The results reveal that AI pilot policies can elevate the level of corporate supply chain resilience (with an average increase of 0.0177 units in supply chain resilience in pilot regions, and a 95% confidence interval of [0.0074–0.0281]). This enhancement is primarily achieved by strengthening enterprises’ absorptive capacity, resource integration capability, and innovation ability, while the digital foundation and capital investment of regions and enterprises further amplify this positive impact. Meanwhile, the policy’s influence exhibits significant heterogeneity, with more pronounced effects in eastern regions, central cities, technology-intensive industries, and state-owned enterprises. When facing external shocks, AI policies can mitigate the adverse impacts caused by such shocks, and this mitigating effect is more significant in the later stages of the shock. Additionally, these policies can drive the improvement of supply chain resilience in non-pilot regions through spatial spillover effects. The conclusions of this study offer practical references for optimizing supply chain management and enhancing supply chain resilience through AI policies, as well as valuable insights for relevant policy formulation and corporate strategic decision-making. © The Author(s) 2025.}, keywords = {Absorptive capacity; Artificial intelligence; Innovation ability; Resource integration capability; Supply chain resilience; absorption; article; artificial intelligence; human; pharmaceutics; pilot study; policy; supply chain}, } @article{enginz2026108, title = {Human-AI governance (HAIG): A trust-utility approach}, author = {Engin Z.}, year = {2026}, journal = {Journal of Responsible Technology}, volume = {26}, pages = {100167}, doi = {10.1016/j.jrt.2026.100167}, abstract = {This paper introduces the Human-AI Governance (HAIG) framework, contributing to the AI Governance (AIG) field by foregrounding the relational dynamics between human and AI actors rather than treating AI systems as objects of governance alone. Current categorical frameworks (e.g., ‘human-in-the-loop’ models) inadequately capture how AI systems evolve from ‘tools’ to ‘partners’, particularly as foundation models demonstrate emergent capabilities and multi-agent systems exhibit autonomous goal-setting behaviours. As systems are deployed across contexts, agency redistributes in complex patterns that are better represented as positions along continua rather than discrete categories. The HAIG framework operates across three levels: dimensions (Decision Authority, Process Autonomy, and Accountability Configuration), continua (continuous positional spectra along each dimension), and thresholds (critical points along the continua where governance requirements shift qualitatively). The framework's dimensional architecture is level-agnostic, applicable from individual deployment decisions and organisational governance through to sectorial comparison and national/international regulatory design. Unlike risk-based or principle-based approaches that treat governance primarily as a constraint on AI deployment, HAIG adopts a trust-utility orientation – reframing governance as the condition under which human-AI collaboration can realise its potential, calibrating oversight to specific relational contexts rather than predetermined categories. The analysis reveals how technical advances in self-supervision, reasoning authority, and distributed decision-making drive context-dependent and non-uniform evolution in trustworthiness conditions — with consequences for the trust attitudes of relevant stakeholders that existing categorical frameworks are ill-equipped to track. Case studies in healthcare and European regulation demonstrate how HAIG complements existing frameworks while offering a foundation for adaptive regulatory design that anticipates governance challenges before they emerge. © 2026 The Author(s). Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license. http://creativecommons.org/licenses/by-nc-nd/4.0/}, keywords = {Agentic AI; AI oversight; Dimensional governance; Foundation models; Human-AI governance (HAIG); Trust dynamics; Trust-utility}, } @article{zulkeplimissafwanharunmikhlasroselemnorazelanshmohdazharmh2026109, title = {ARTIFICIAL INTELLIGENCE IN ISLAMIC FINANCE: AN OUTLOOK BASED ON MAQASID AL-SHARIAH}, author = {Zulkepli M.I.S. and Safwan Harun M. and Ikhlas Rosele M. and Norazelan S.H. and Mohd Azhar M.H.}, year = {2026}, journal = {Journal of Fatwa Management and Research}, volume = {31}, number = {2}, pages = {304-335}, doi = {10.33102/jfatwa.vol31no2.780}, abstract = {The advancement of Artificial Intelligence (AI) has significantly influenced global financial systems, including the Islamic finance sector. AI technologies such as robo-advisors, RegTech, and automated Shariah screening systems have enhanced efficiency and decision-making. However, their application requires critical evaluation to ensure alignment with Islamic ethical and legal principles. Despite the growing adoption of AI in Islamic finance, limited studies have explored its implementation from the perspective of maqāṣid al-Sharī‘ah (objectives of Islamic law). The absence of a clear ethical and jurisprudential framework raises concerns such as transparency, algorithmic bias, and Shariah compliance. This study aims to analyse the application of AI in Islamic finance from the perspective of maqāṣid al-Sharī‘ah, and propose a maqāṣid-based strategy for AI operations in Islamic finance. This study employs a qualitative research design based on library and interpretive analysis of classical and contemporary Shariah texts, peer-reviewed academic literature, regulatory guidelines, and institutional governance frameworks related to Islamic finance and artificial intelligence. This study found that AI can support the realization of maqāṣid al-Sharī‘ah by promoting transparency, justice, and public welfare in Islamic finance operations. Applications such as AI-driven credit scoring, robo-advisory services, and compliance automation contribute to financial inclusion, equitable wealth distribution, and risk mitigation. Nonetheless, ethical challenges such as privacy violations, data misuse, and opacity in algorithmic decision-making must be carefully addressed to prevent contradictions with Shariah principles. This study contributes to the emerging discourse on ethical AI integration within Islamic finance by proposing a maqāṣid al-Sharī‘ah-based evaluative framework. It highlights the importance of developing Shariah-oriented AI governance and ethical policies to guide Islamic financial institutions and regulators. The findings provide a foundation for future empirical research and policy formulation aimed at harmonizing technological innovation with Islamic ethical objectives. © The Authors 2026.}, keywords = {Artificial intelligence; Islamic finance; Maqāṣid al-Sharī‘ah; Shariah compliance}, } @article{pigactleeahuanga2026110, title = {Navigating Transparency in AI-Powered Luxury Hospitality: A Dynamic Guest-Centric Approach}, author = {Pigac T. and Lee A. and Huang A.}, year = {2026}, journal = {Cornell Hospitality Quarterly}, volume = {67}, number = {3}, pages = {283-297}, doi = {10.1177/19389655261433944}, abstract = {This study explores how artificial intelligence (AI) transparency can be designed to enhance trust and guest experience in luxury hospitality. Drawing on 50 semi-structured interviews with hotel guests across Europe, Asia, and North America, and segmented using the CEW Technology Comfort Scale, the research develops the Dynamic Transparency Protocol (DTP) framework. Findings reveal that transparency preferences vary across guest profiles and service stages, shaped by three adaptive mechanisms: user-centric adaptation, situational sensitivity, and emotional matching. Guests with lower digital comfort valued human-mediated, simplified disclosures, while digital elites demanded customizable dashboards and traceability. Across segments, emotional resonance emerged as critical for perceived fairness and trust, reframing transparency as both informational and affective. The study contributes by contextualizing transparency and trust frameworks in a luxury service setting and offers actionable guidance for managers on tiered transparency design, emotionally tuned interfaces, and hybrid human–AI mediation. © The Author(s) 2026}, keywords = {artificial intelligence; guest experience; luxury hospitality; personalization; transparency; trust}, } @article{jaturatnnanankhub2026111, title = {Measuring AI responsibility: A cross-country validation of a multidimensional framework}, author = {Jaturat N. and Na-Nan K. and Hu B.}, year = {2026}, journal = {International Journal of Information Management Data Insights}, volume = {6}, number = {1}, pages = {100388}, doi = {10.1016/j.jjimei.2026.100388}, abstract = {As artificial intelligence (AI) continues to transform industries, ensuring AI responsibility has become critical for ethical governance. Despite the growing number of frameworks emphasizing transparency, accountability, and sustainability, a standardized measurement tool remains lacking. This study develops and validates a seven-dimensional AI Responsibility framework encompassing Privacy and Security, Transparency and Accountability, Impact on Employment, Sustainability, User-Centered Design, Social Impact, and Innovation and Adaptation. Using Exploratory Factor Analysis (EFA) and Confirmatory Factor Analysis (CFA), the study confirms the framework’s construct validity and reliability. The results indicate strong model fit, with all constructs exceeding recommended thresholds for composite reliability (CR) and average variance extracted (AVE). The study contributes to AI ethics research by offering an empirically validated measurement instrument. Practically, the framework serves as a benchmarking tool for organizations and policymakers to assess AI governance strategies and regulatory compliance. As AI adoption continues to expand, this framework provides a structured approach to fostering trust, accountability, and responsible AI deployment. © 2026 The Author(s).}, keywords = {Accountability; Ai governance; Ai responsibility; ethical AI; innovation; sustainability; Transparency}, } @article{peitchenxzhengl2026112, title = {Artificial intelligence policy and supply chain disruption risk: Evidence from innovation-driven and cost-saving effects}, author = {Pei T. and Chen X. and Zheng L.}, year = {2026}, journal = {Economic Modelling}, volume = {160}, pages = {107593}, doi = {10.1016/j.econmod.2026.107593}, abstract = {This study investigates the impact of Artificial Intelligence (AI) policy on firms’ supply chain disruption risk. Prior research has focused on the effect of AI technology on supply chain risk, while paying less attention to the role of AI-related policies. Using data on Chinese listed firms from 2012 to 2024 and exploiting the implementation of China’s National AI Innovative and Application Pilot Zones as a quasi-natural experiment, we find that AI policy significantly reduces firms’ supply chain disruption risk. Our analysis reveals that the risk-mitigating effect operates through higher innovation investment and lower operating costs, and the aboveeffect is more pronounced for non-state-owned enterprises, smaller firms, and firms with more concentrated supply chains. Further analysis indicates that the risk-mitigating effect of AI policy generates spillovers along supply chains. This study offers practical insights for policy design to integrate technological progress with supply chain risk management. © 2026 Elsevier B.V.}, keywords = {Artificial intelligence innovation and application pilot zones; Artificial intelligence policy; Spillover effects; Supply chain disruption risk}, } @article{nandagopals2026113, title = {Achieving Information Equilibrium: An Integrated Economic-Sociotechnical-Institutional Framework for Governing Artificial Intelligence Asymmetries}, author = {Nandagopal S.}, year = {2026}, journal = {IEEE Transactions on Technology and Society}, volume = {7}, number = {2}, pages = {133-141}, doi = {10.1109/TTS.2026.3680525}, abstract = {Artificial intelligence systems now scores credit, routes trucks, and curates news, yet their decisions remain largely inscrutable to users, regulators, and even the firms that deploy them. Such opacity creates information asymmetries that invite principal-agent conflicts, adverse selection, and moral hazard, undermining both market efficiency and public trust. Existing remedies are fragmented: information-economics studies prescribe incentives but overlook design; sociotechnical research clarifies human-AI interaction but neglects market power; and institutional work codifies norms without detailing engineering trade-offs. This paper weaves those strands into a single AI Information-Equilibrium Framework (AIIEF). The framework integrates three governance layers. At the micro layer, explainable dashboards, model cards, and user-training loops reveal how an algorithm works and calibrate reliance. At the meso layer, liability clauses, reciprocal data trusts, and tiered transparency labels align incentives so that actors share and use information truthfully. At the macro layer, risk-tiered regulation, auditable standards such as ISO/IEC 42001, and global principles issued by the OECD and UNESCO transform ethical aspirations into enforceable obligations. I show how AIIEF adapts across AI modalities - agentic versus tool-like, large versus small models - and deployment contexts - enterprise versus consumer. I also address common objections - gaming, innovation chill, free-riding, and superficial oversight - by coupling tiered disclosure with regulatory sandboxes, data-trust reciprocity, and decision-right-calibrated explanations. Finally, I outline a research agenda on asymmetry metrics, longitudinal capability evolution, ecosystem simulations, and institutional diffusion, positioning AIIEF as both a cumulative theory for scholars and a pragmatic blueprint for managers. © 2020 IEEE.}, keywords = {AI governance; algorithmic management; incentives; information asymmetry; institutional regulation; sociotechnical systems; transparency; trust; Artificial intelligence; Commerce; Ecosystems; Ethical aspects; Human engineering; Information management; Information systems; ISO Standards; Product liability; AI governance; Algorithmic management; Algorithmics; Incentive; Information asymmetry; Information equilibrium; Institutional regulation; Sociotechnical; Sociotechnical systems; Trust; Economic and social effects}, } @article{papyshevgchankjdmigliorinis2026114, title = {AI Risk Bonds: a market-based mechanism for governing liability}, author = {Papyshev G. and Chan K.J.D. and Migliorini S.}, year = {2026}, journal = {Data and Policy}, volume = {8}, pages = {e27}, doi = {10.1017/dap.2026.10079}, abstract = {The rapid proliferation of AI systems has outpaced regulatory and insurance frameworks, leaving risks from unpredictable rogue AI behaviors unaddressed. While academic debates prioritize existential threats, this article shifts focus to governing present-day AI through AI Risk Bonds: market-driven instruments inspired by catastrophe bonds. These bonds securitize AI-related liabilities, using investor scrutiny to price risks based on a system’s expected impact and behavioral predictability. By dynamically adjusting bond yields, higher risks escalate capital costs for developers, incentivizing proactive risk mitigation. The mechanism addresses regulatory blind spots via market oversight, disperses liability through capital markets, and reduces moral hazard by linking financing to risk profiles. Complementing initiatives like the EU AI Act, this framework balances innovation with precaution, tethering profitability to risk minimization for responsible AI development. © The Author(s), 2026. Published by Cambridge University Press.}, keywords = {AI liability; AI risks; catastrophe bonds; market-based mechanism; risk sharing}, } @article{aliatariqh2026115, title = {Do intelligent HR technologies spark or stifle employee creativity? A dual-pathway perspective}, author = {Ali A. and Tariq H.}, year = {2026}, journal = {Technology in Society}, volume = {88}, pages = {103433}, doi = {10.1016/j.techsoc.2026.103433}, abstract = {Artificial intelligence is increasingly embedded in human resource management, yet we know relatively little about when and how intelligent HR technologies shape employee creativity. Drawing on conservation of resources theory and uncertainty reduction theory, we develop a dual-pathway model in which intelligent HR technology use can build resources by enhancing psychological empowerment while also depleting resources by triggering rumination. We further examine AI self-efficacy as a boundary condition and transparency and trust as uncertainty-reducing antecedents of technology use. We test the model across two multi-wave, multi-source studies in China. Study 1 examines algorithmic HRM in a large manufacturing organization and uses HR-provided creativity reward categories as an indicator of creativity (N = 529). Study 2 examines AI-enabled HR tools among HR professionals and uses leader-rated creativity (N = 245). Across both studies, intelligent HR technology use was positively associated with both psychological empowerment and rumination; empowerment, in turn, was positively related to creativity, whereas rumination was negatively related to creativity. AI self-efficacy strengthened the empowering pathway and attenuated the ruminative pathway. In Study 2, transparency and trust were positively related to employees’ use of intelligent technologies in HRM. These findings clarify why the same class of HR technologies can generate both creativity-relevant benefits and costs. They further suggest that responsible implementation requires transparent and trustworthy AI governance, alongside capability-building efforts that reduce rumination while sustaining empowerment. Copyright © 2026. Published by Elsevier Ltd.}, keywords = {AI self-efficacy; AI-Enabled HR tools; Algorithmic HRM; Employee creativity; Intelligent technologies in HRM; Psychological empowerment; Rumination; Transparency; Trust; Conservation; Copyrights; Empowerment of personnel; Human resource management; Natural resources management; AI self-efficacy; AI-enabled HR tool; Algorithmic HRM; Algorithmics; Employee creativities; Intelligent technology; Intelligent technology in HRM; Psychological empowerments; Rumination; Self efficacy; Trust; Transparency}, } @article{alyasiriay2026116, title = {The integration of artificial intelligence in radiation medical physics: insights from an international survey with regional variability}, author = {Al-Yasiri A.Y.}, year = {2026}, journal = {Journal of Radiological Protection}, volume = {46}, number = {2}, pages = {021517}, doi = {10.1088/1361-6498/ae6d09}, abstract = {Artificial intelligence (AI) is considered to be a leading technology in radiation medical physics, which has the potential for improving efficiency and precision in imaging, radiotherapy, and nuclear medicine. Nonetheless, its application in the clinical setting is hindered by education, regulation, and ethical issues. To evaluate the views of medical physicists on AI adoption, potential issues, perceived advantages, ethical issues, and training needs. A cross-sectional survey in the form of an online questionnaire was distributed internationally to practicing medical physicists, representing specialties in radiation therapy, diagnostic imaging, and nuclear medicine. The instrument was used to capture demographic variables and AI familiarity, perceived benefits and barriers, ethical issues, training preferences, and future expectations. Descriptive statistics, chi-square tests, and binary logistic regression were performed. Open-ended responses were analyzed using thematic analysis. Most respondents were moderately familiar with AI, but nearly half had not incorporated AI into their practice despite being interested. Adoption patterns differed significantly across geographic regions, with higher usage reported in developed countries. AI familiarity was strongly associated with adoption. The most prominent benefits noted were dose calculation accuracy, patient safety, and workflow efficiency. In contrast, the main obstacles were inadequate training, high implementation costs, absence of standardized protocols, and low availability of advanced equipment. Ethical apprehensions focused on accountability and reduced human oversight. Preferred educational strategies included hands-on workshops and on-the-job training. Regional analysis and modeling revealed variability in AI adoption and identified key predictors, including AI familiarity and specialization. Results indicate that AI is considered an important technology in radiation medical physics. However, implementation was influenced by regional and professional differences. These findings highlight the necessity to expand AI education, create standard training programs, improve infrastructure, and provide clearer governance frameworks to support safe and responsible AI integration. © 2026 Society for Radiological Protection. Published on behalf of SRP by IOP Publishing Limited. All rights, including for text and data mining, AI training, and similar technologies, are reserved. This article is available under the terms of the https://publishingsupport.iopscience.iop.org/iop-standard/v1.}, keywords = {artificial intelligence; clinical integration; radiation medical physics; the viewpoints of medical physicists; Artificial Intelligence; Cross-Sectional Studies; Female; Health Physics; Humans; Male; Surveys and Questionnaires; Ethical aspects; Integration; Logistic regression; Medical imaging; Nuclear medicine; Personnel training; Population statistics; Radiotherapy; Regional planning; Statistical tests; Clinical integration; Ethical issues; Improving efficiency; International survey; Leading technology; Medical physicists; Medical physics; Radiation medical physic; Regional variability; The viewpoint of medical physicist; aged; article; artificial intelligence; chi square distribution; data mining; developed country; diagnostic imaging; dose calculation; female; human; in service training; logistic regression analysis; medical physicist; middle aged; nuclear medicine; online questionnaire; patient safety; radiotherapy; responsible artificial intelligence; thematic analysis; workflow; workshop; cross-sectional study; health physics; male; questionnaire; Diagnosis}, } @article{goyalrdeshmukhsgbolian2026117, title = {Navigating barriers to GenAI adoption in public administration: A systematic evaluation and policy roadmap}, author = {Goyal R. and Deshmukh S.G. and Bolia N.}, year = {2026}, journal = {Socio-Economic Planning Sciences}, volume = {105}, pages = {102461}, doi = {10.1016/j.seps.2026.102461}, abstract = {Public administration serves as a crucial pillar of governance, ensuring efficient policy implementation, resource management, and service delivery. However, administrative processes often face challenges such as inefficiencies, resource constraints, and bureaucratic delays. The emergence of Generative Artificial Intelligence (GenAI) presents an opportunity to enhance decision making, optimize administrative workflows, and improve public service efficiency. Despite its potential, the widespread adoption of GenAI in public administration is prevented by various barrier, thus a systematic assessment and prioritizing is necessary to facilitate structured implementation. This study aims to identify, analyze, and categorize key barriers to GenAI adoption. It begins with a comprehensive literature review to identify potential barriers. Utilizing the Technology-Organization-Environment (TOE) framework, the study systematically classifies these barriers and employs the Fuzzy Best-Worst Method (FBWM) and Decision-Making Trial and Evaluation Laboratory (DEMATEL) to prioritize them based on their significance, causal influence and interdependencies. The novelty of this study lies in its structured segmentation of barriers offering a comprehensive understanding of the challenges impeding GenAI adoption. The study presents a GenAI use case in India, highlighting its practical application in one of the world's largest administrative ecosystems. Given India's complex governance structure, addressing these barriers provides a scalable model for Artificial Intelligence (AI) driven administration globally. Additionally, study proposes time bound policy recommendations for the systematic integration of GenAI in public administration, ensuring accountability and ethical compliance. This research also contributes to academic discourse and policy formulation, emphasizing the need for continuous AI policy assessment to keep pace with technological advancements and evolving governance structures. © 2026 Elsevier Ltd.}, keywords = {DEMATEL; FBWM; GenAI; Public administration; Smart governance; TOE; India; artificial intelligence; governance approach; policy implementation; public administration; public service; technology adoption}, } @article{shenp2026118, title = {Toward a proportionate explainability framework in administrative automated decision-making: A lesson from China}, author = {Shen P.}, year = {2026}, journal = {Computer Law and Security Review}, volume = {61}, pages = {106307}, doi = {10.1016/j.clsr.2026.106307}, abstract = {While explainability has become a central concern in AI governance, the specific challenges surrounding explainability in administrative decisions generated by automated decision-making (ADM) systems remain underexplored. Across global administrative law enforcement practices, governments normally bear the duty to give reasons for their decisions, but they often fail to provide meaningful explanations to affected individuals, either offering excessive technical detail or failing to disclose sufficient reasoning in the context of automated decision-making. The unsystematic and fragmented nature of the explainability framework further exacerbates the problem structurally. This article explores how to establish a robust framework for achieving effective explainability in the context of administrative automated decision-making. Using China as a case study, the analysis of Chinese administrative practices reveals persistent shortcomings, such as disclosure without explanation and inadequate substantive reasoning. These issues largely stem from an imbalance among three interrelated principles: accountability, efficiency, and procedural fairness. In particular, an excessive emphasis on efficiency has undermined both accountability and fairness in the implementation of explainability mechanisms. Building on these findings, a proportionate explainability framework for administrative automated decision-making should be established. The framework introduces a dual-level model, one for external oversight bodies and another for affected individuals, each bound by the principle of scaling explanation depth with the intensity of administrative power exercised. This approach seeks to maintain a dynamic equilibrium among the three interrelated principles of accountability, efficiency, and procedural fairness. In doing so, it aims to promote effective explainability by balancing the interests of the key actors involved: the government as decision-maker, supervisory institutions as oversight bodies, and individuals as the subjects of administrative decisions. © 2026 The Author(s)}, keywords = {Accountability; Administrative law; Automated decision making; Efficiency; Procedural fairness; Proportionate explainability; Automation; Decision making; Information systems; Information use; Law enforcement; Accountability; Administrative law; Automated decision making; Automated decision making systems; Case-studies; Enforcement practices; Level model; Procedural fairness; Proportionate explainability; Technical details; Efficiency}, } @article{rashidmt2026119, title = {From Recommender to Actor: The Normative Boundary When RAG Tools Become Tool-Calling Agents}, author = {Rashid M.T.}, year = {2026}, journal = {Minds and Machines}, volume = {36}, number = {2}, pages = {30}, doi = {10.1007/s11023-026-09782-z}, abstract = {Retrieval-Augmented Generation (RAG) systems increasingly operate not only as tools for retrieving and synthesizing information, but also as agents that can invoke external functions, modify digital environments, and execute tasks across software systems. This development raises a specific normative problem: the point at which a model’s output ceases to be merely informational and becomes an executable intervention in the world. Building on existing work in Responsible AI, accountability, and human oversight, this paper argues that tool-calling architectures place particular pressure on these frameworks because they can fuse retrieval, reasoning, and action within a single operational pipeline. To clarify this transition, the paper develops a threshold account of performative action based on four criteria: causal efficacy, autonomy, irreversibility, and moral salience. It then examines the normative consequences of crossing this boundary, showing how executable outputs can fragment responsibility, weaken effective oversight, and produce miscalibrated trust in hybrid human-AI systems. In response, the paper proposes three governance-by-design heuristics for tool-calling environments: epistemic traceability, operational reversibility, and normative containment. Together, these mechanisms aim to make actionable systems more answerable by preserving visibility into how actions arise, by creating limited opportunities for interruption or correction, and by bounding the scope of permissible delegation. The paper concludes that reclaiming the boundary between knowing and doing is essential not to restrict intelligence, but to preserve the conditions under which increasingly capable AI systems remain governable within human practices of authorization, accountability, and repair. © The Author(s) 2026.}, keywords = {AI ethics; Distributed agency; Governance by design; Normative boundary; Retrieval-augmented generation; Tool-calling agents}, } @article{chuwwlamjlc2026120, title = {Transforming healthcare management: The impact of artificial intelligence on leadership and operations}, author = {Chu W.W. and Lam J.L.C.}, year = {2026}, journal = {Management in Healthcare}, volume = {10}, number = {3}, pages = {253-265}, doi = {10.69554/AQMU2524}, abstract = {Artificial intelligence (AI) is reshaping healthcare leadership by combining technological innovation with a focus on human values. This paper examines how AI helps tackle workforce shortages, cost pressures and inequalities. The integration of AI into health care is transforming how services are delivered and managed. As healthcare systems worldwide face significant challenges, such as an ageing population, increasing rates of chronic diseases and rising patient expectations. AI offers innovative solutions to improve efficiency and patient care. In many regions, healthcare providers are overwhelmed, struggling to meet the growing demand for quality services. AI technologies, such as machine learning and natural language processing, are being used to enhance patient engagement, streamline administrative tasks and improve clinical decision making. These tools can analyse large amounts of data to provide personalised care and support, helping healthcare professionals make better-informed decisions. For instance, AI can assist in developing personalised treatment plans, enabling providers to address individual patient needs more effectively. Moreover, AI can automate routine tasks, allowing healthcare staff to focus on more complex and value-added activities, ultimately enhancing the quality of care. Nevertheless, the adoption of AI in healthcare is not without its challenges. Concerns about data privacy, algorithmic bias, and the need for ethical guidelines must be carefully managed to ensure that AI solutions are fair and safe for all patients. By 2030, the vision for AI in health care is to create a more efficient, accessible and patient-centred system. This paper explores the impact of AI on healthcare management, highlighting the opportunities it presents while addressing the necessary considerations for successful implementation. This article is also included in The Business & Management Collection which can be accessed at https://hstalks.com/business/. Delivered by Ingenta © Henry Stewart Publications 2397-1061 (2026).}, keywords = {artificial intelligence; ethical leadership; healthcare management; operational efficiency; predictive analytics}, } @article{oliveiradnunessbregofnunesr2026121, title = {Ethical–Regulatory Guidelines for AI in Palliative Care Rehabilitation}, author = {Oliveira D. and Nunes S.B. and Rego F. and Nunes R.}, year = {2026}, journal = {Healthcare (Switzerland)}, volume = {14}, number = {7}, pages = {895}, doi = {10.3390/healthcare14070895}, abstract = {Background/Objectives: The integration of artificial intelligence (AI) into rehabilitation practice has expanded rapidly, including its emerging application in palliative care contexts. Although international organisations have established ethical and governance frameworks for AI in healthcare, these initiatives remain largely high-level and are not specifically tailored to the clinical complexity, vulnerability, and relational dimensions of palliative care rehabilitation. The absence of context-specific ethical–regulatory guidance poses challenges for responsible implementation in ethically sensitive settings. This study aimed to consolidate ethically grounded regulatory guidance for the use of AI in palliative care rehabilitation by translating existing international principles into context-sensitive domains. Methods: A qualitative documentary analysis with a normative ethical–regulatory orientation was conducted using the READ (Ready, Extract, Analyse, Distil) framework. Authoritative international policy, governance, and regulatory documents addressing AI in healthcare were identified and analysed. Data were extracted using a structured analytical table and coded according to predefined ethical–regulatory domains derived from previously published ethical guidelines and verified through documentary analysis. Results: The analysis identified five convergent ethical–regulatory domains recurrent across international governance frameworks: (1) Human oversight and clinical responsibility; (2) Patient autonomy, preferences, and proportionality; (3) Transparency and explainability; (4) Fairness, equity, and non-discrimination; and (5) Professional competence and ethical literacy. These domains were synthesised into practical ethical–regulatory considerations linking ethical principles with governance expectations and clinical implementation requirements. Conclusions: This study provides context-sensitive ethical–regulatory guidance that bridges high-level AI governance principles with the operational realities of palliative care rehabilitation. By systematising and operationalising existing ethical norms, the proposed framework supports responsible clinical decision-making, strengthens institutional accountability, and safeguards patient dignity and autonomy in vulnerable care contexts. © 2026 by the authors.}, keywords = {AI governance; artificial intelligence; clinical decision support; ethical–regulatory guidance; palliative care rehabilitation; patient-centred care; article; artificial intelligence; clinical decision making; clinical decision support system; clinical practice guideline; fairness; human; human dignity; international organization; palliative therapy; patient autonomy; person centered care; practice guideline; professional competence; rehabilitation; vulnerability}, } @article{sayyadim2026122, title = {The ethics of AI: Pursuing accountability for assured data to make right decisions}, author = {Sayyadi M.}, year = {2026}, journal = {Business Information Review}, volume = {43}, number = {1}, pages = {14-19}, doi = {10.1177/02663821261427241}, abstract = {AI disruptions will bring vast benefits and challenges to companies. One key question remains: How can companies overcome corporate accountability challenges in the AI age? To answer this question, the article explores how to assign accountability when artificial intelligence systems are involved in decision-making. As AI becomes more widespread, who should be held responsible if these systems make poor choices is unclear. The traditional top-down accountability model, from executives to managers, faces challenges due to AI’s black-box nature. Approaches such as holding developers or users liable have limitations as well. It is argued that shared accountability across multiple stakeholders may be optimal but supported by testing, oversight committees, guidelines, regulations, and explainable AI Concrete finance, customer service, and surveillance examples illustrate AI accountability issues. The paper summarizes perspectives from academia and business practice on executives’ and boards’ roles, including mandating audits and transparency. It concludes that while AI accountability models remain debated, decision-makers must take responsibility for the technologies deployed. The article suggests combining prescriptive accountability rules and data quality evaluation frameworks can optimize resources to enhance AI-assisted decision-making, align regulatory requirements, respect stakeholders, and exploit competitive advantage using advanced technology. © The Author(s) 2026}, keywords = {accountability; artificial intelligence; customer service; data; decision-making; ethics; security}, } @article{agarwalasebastianmpkrishnans2026123, title = {Understanding GenAI Teammates in the Workplace: A Sensemaking and Sensegiving Analysis of User Reviews}, author = {Agarwal A. and Sebastian M.P. and Krishnan S.}, year = {2026}, journal = {Information Systems Frontiers}, volume = {28}, number = {2}, pages = {803-825}, doi = {10.1007/s10796-025-10672-5}, abstract = {Generative AI (GenAI) applications, such as ChatGPT, are increasingly shaping work practices and employee engagement in organizations. Understanding how employees interact with these tools is critical for designing effective and responsible AI-enabled workplaces. This study analyzes 443,338 user reviews from the Google Play Store to examine how GenAI tools influence user satisfaction, continued use and their behaviors, which in turn impact productivity and well-being. Drawing on Sensemaking and Sensegiving theories, we develop a four-stage framework integrated into a 3E model (Envision-Evolve-Engage) comprising seven propositions. Findings highlight GenAI’s potential to enhance workplace effectiveness, decision-making and employee well-being, and to advance Sustainable Development Goal 8 (SDG 8) by promoting productive, inclusive, and meaningful work. The study also identifies challenges related to trust, privacy, adaptability, and ethical use. These insights offer practical guidance for designing user-centric GenAI systems and provide a theory-driven perspective for supporting responsible adoption and engagement in workplace contexts. © The Author(s) 2026.}, keywords = {AI teammates; Big data analytics; Conversational AI; Employee productivity; Employee well-being; Future of work; Generative AI; Text mining; User review; Workplace innovation; Advanced Analytics; Artificial intelligence; Behavioral research; Data mining; Decision making; Personnel; Sustainable development; AI teammate; Big data analytic; Conversational AI; Data analytics; Employee productivity; Employee well-being; Future of works; Generative AI; Text-mining; User reviews; Well being; Workplace innovation; Big data}, } @article{xiahchenhzhangjzkamalmm2026124, title = {Exploring the impact of responsible AI governance on corporate performance: A quasi-natural experiment}, author = {Xia H. and Chen H. and ZHANG J.Z. and KAMAL M.M.}, year = {2026}, journal = {Technological Forecasting and Social Change}, volume = {223}, pages = {124425}, doi = {10.1016/j.techfore.2025.124425}, abstract = {The rapid advancement of artificial intelligence (AI) presents significant ethical, legal, and operational challenges, elevating the importance of responsible AI governance. This paper examines the impact of responsible AI information governance mechanisms on corporate performance using a sample of 342 securities firms. Companies are categorized into control and experimental groups based on their responsible AI application scores. Treating the adoption of responsible AI governance practices as a “quasi-natural” experiment, the study applies a multi-period difference-in-differences (DID) regression model to uncover insights into causal relationships. The study also investigates the relative importance of various responsible AI governance dimensions in driving performance improvements. The results demonstrate that firms adopting responsible AI governance practices see significant performance improvements. This study adds to the academic discourse by deepening understanding of how responsible AI governance influences business outcomes and by offering a theoretical foundation for future research. From a practical perspective, it provides actionable guidance for organizations, demonstrating how ethical AI governance can drive sustainable business growth and enhance overall corporate performance. © 2025 The Authors}, keywords = {Corporate performance; Difference-in-differences (DID); Information governance mechanisms; Responsible AI; Artificial intelligence; Ethical technology; Industrial management; Control groups; Corporate performance; Difference-in-difference; Difference-in-differences; Differences-in-differences; Governance mechanisms; Information governance mechanism; Natural experiment; Operational challenges; Responsible artificial intelligence; artificial intelligence; corporate strategy; experimental study; future prospect; governance approach; Regression analysis}, } @article{spieglershodarpanta2026125, title = {Images of AI: How AI practitioners view the impact of Artificial Intelligence on society, now and in the future}, author = {Spiegler S. and Hoda R. and Pant A.}, year = {2026}, journal = {Technology in Society}, volume = {84}, pages = {103109}, doi = {10.1016/j.techsoc.2025.103109}, abstract = {Despite unprecedented technological advancement, intense commercial investment, international agreements, and growing societal concerns with Artificial Intelligence (AI), there is little insight into how those driving the field – the everyday AI practitioners – perceive AI and its impact on society, now and in the future. We address this critical gap by conducting a broad-based survey with 100 AI practitioners, followed by in-depth interviews with 20 AI practitioners, including developers, managers, and consultants. Using socio-technical grounded theory (STGT) for data analysis, we inductively identified six images of AI which capture six ways in which AI practitioners view AI, now and in the future, and their implications for impact on society and human control : Parrot captures AI that mimics human behaviour, including biases; Companion surrounds humans in daily life and supports decision making with empathy-like traits; Wolf in Sheep’s Clothing highlights AI misused by humans, causing societal harms; Saviour envisions AI solving complex problems beyond human capacity; Wizard portrays AI as powerful, yet, unpredictable and inexplicable; and Pinocchio imagines AI as gaining free will, learning from mistakes, and possibly harming humans. These images of AI provide a novel framework for understanding how AI practitioners perceive and shape AI solutions. Our findings and recommendations will assist AI practitioners, companies, and users with a shared vocabulary and understanding to explicitly and critically examine the intended and unintended impacts of AI solutions on human society, contributing to more responsible and human controlled AI design and use. © 2025 The Author(s).}, keywords = {Agentic AI; AI; AI practitioners; AI psychosis; Artificial intelligence; Future; Generative AI; Human control; Images; Large language models; LLMs; Responsible AI; Societal impact; Behavioral research; Decision making; Decision theory; Economic and social effects; Human engineering; Investments; Man machine systems; Agentic artificial intelligence; Artificial intelligence practitioner; Artificial intelligence psychosis; Future; Generative artificial intelligence; Human control; Image; Language model; Large language model; LLM; Responsible artificial intelligence; Societal impacts; artificial intelligence; future prospect; human behavior; machine learning; social impact; survey; Artificial intelligence}, } @article{jaciowmadamczykabartukbratnickamyliwieckhoffmannburdziskakskrskaastrzeleckiaszojdagwolnyr2026126, title = {Perceived Benefits, Leadership Engagement and AI Maturity in Polish SMEs: A Socio-Technical Perspective on Sustainable Digital Transformation Under Competitive Pressure}, author = {Jaciow M. and Adamczyk A. and Bartuś K. and Bratnicka-Myśliwiec K. and Hoffmann-Burdzińska K. and Skórska A. and Strzelecki A. and Szojda G. and Wolny R.}, year = {2026}, journal = {Sustainability (Switzerland)}, volume = {18}, number = {10}, pages = {4807}, doi = {10.3390/su18104807}, abstract = {Digitalization and artificial intelligence (AI) are seen as promising pathways for small and medium-sized enterprises (SMEs) to enhance performance while preserving environmental and social resources. This paper identifies organizational determinants of AI maturity that can enable SMEs to use AI in a more sustainable, responsible, and capacity-enhancing manner. AI adoption becomes relevant to sustainability not only because a company adopts advanced technology but because this technology is embedded in leadership practices, employee competencies, interdisciplinary collaboration, and organizational learning. From this perspective, perceived benefits and management commitment are not outcomes of sustainability but mechanisms that help explain how SMEs transition from technological awareness to building organizational capacity. Such capacity building can be a necessary prerequisite for subsequent sustainability-oriented outcomes, such as efficient resource utilization, employee upskilling, responsible AI management, and long-term resilience. We conducted a cross-sectional survey among 402 managers from Polish SMEs (62 micro, 193 small, 147 medium) across manufacturing, services and trade industries. Respondents (mean age ≈ 42.5 years) assessed perceived benefits of AI, engagement of top leadership, AI maturity and competitive pressure. Partial least-squares structural equation modeling revealed that perceived benefits strongly predicted leadership engagement (β = 0.647), explaining 62.8% of its variance. Perceived benefits (β = 0.384) and leadership engagement (β = 0.362) in turn were the key drivers of AI maturity, with the model accounting for 65.5% of variance in AI maturity. Competitive pressure positively but weakly moderated the relationship between perceived benefits and leadership engagement (β = 0.011), while its moderating effect on the relationship between perceived benefits and AI maturity was not significant (β = −0.008). These findings suggest that articulating clear benefits of AI and securing active leadership engagement are more decisive for advancing AI maturity than external competitive pressure. The contribution of the study is to integrate the perceived benefits of AI, top management commitment and AI maturity into a model, empirically validated and interpreted from a socio-technical perspective of sustainable digital transformation in SMEs, while quantifying the moderating role of competitive pressure in the under-researched context of Central and Eastern Europe. For practitioners, investing in awareness of AI’s benefits and developing committed leadership may yield more sustainable digital transformation than reacting solely to external pressures. © 2026 by the authors.}, keywords = {AI maturity; CAWI; leadership engagement; perceived benefits; Poland; SME; sustainable digital transformation}, } @article{huynhgnguyenhtnvoltevansevuongnmanpl2026127, title = {From COVID-19 to Nipah virus preparedness in low- and middle-income countries: Re-centering points of decision and responsible AI}, author = {Huynh G. and Nguyen H.T.N. and Vo L.T. and Evans E. and Vuong N.M. and An P.L.}, year = {2026}, journal = {Asian Pacific Journal of Tropical Medicine}, volume = {19}, number = {4}, pages = {145-147}, doi = {10.4103/apjtm.apjtm_118_26}, abstract = {[No abstract available]}, } @article{nelsona2026128, title = {Structured Absence}, author = {Nelson A.}, year = {2026}, journal = {Social Media and Society}, volume = {12}, number = {2}, doi = {10.1177/20563051261452539}, abstract = {This essay argues that while generative artificial intelligence (AI) can be described as a platform, that description is analytically insufficient for understanding where power operates and how it displaces democratic accountability. The platform concept directs attention to the interface—application programming interfaces (APIs), app stores, and developer ecosystems—while the decisive conditions of generative AI lie upstream in compute supply chains, mineral extraction, semiconductor chokepoints, energy and water infrastructures, immigration flows, and corporate-state arrangements that concentrate control. To name this formation, I introduce the concept of structured absence: the organized diminution of democratic institutional capacity in which governance persists while public authorization disappears. Generative AI is not ungoverned; it is extensively governed through export controls, industrial policy, labor regimes, and private infrastructures that remain largely inaccessible to democratic publics. I argue that “predatory inclusion” and authorization are central analytic terms for understanding this shift: communities are incorporated into AI systems on unequal and opaque terms, while the standing to deliberate over those systems is withheld. The essay concludes by calling for democratic infrastructures capable of authorizing technological change. © The Author(s) 2026. This article is distributed under the terms of the Creative Commons Attribution-NonCommercial 4.0 License (https://creativecommons.org/licenses/by-nc/4.0/) which permits non-commercial use, reproduction and distribution of the work without further permission provided the original work is attributed as specified on the SAGE and Open Access pages (https://us.sagepub.com/en-us/nam/open-access-at-sage).}, keywords = {AI governance; authorization; generative AI; platform; platform studies; predatory inclusion; social media; structured absence}, } @article{singhapandeyj2026129, title = {Artificial intelligence applications in human resource management: it is a mixed bag!}, author = {Singh A. and Pandey J.}, year = {2026}, journal = {International Journal of Productivity and Performance Management}, volume = {75}, number = {4}, pages = {1072-1091}, doi = {10.1108/IJPPM-09-2024-0599}, abstract = {Purpose – Artificial intelligence (AI) has brought major disruptions in the new generation human resources (HR) ecosystems. The research community as well as chief human resources officers (CHROs) have been taking strong initiatives to examine the use of AI in human resource management (HRM) function, including harmonious human–machine collaboration. The AI-HRM area is under-researched, and this study addresses an important research gap regarding the benefits and challenges of AI applications in HRM. Design/methodology/approach – The study adopts a qualitative research methodology (abductive case research) and collects data from multiple sources in three Indian companies. These organizations span diverse sectors and were at different stages of AI adoption in HRM at the time of the study. The multi-data-sources strategy helps triangulation and establishes credibility of the research. Findings – The findings provide a clear view about the benefits of AI applications in HRM, higher productivity, recruitment efficiency, adaptive learning and high-quality HR decisions. The study also underpins key challenges, including a lack of human touch, employees’ loss of control over jobs and the fear of losing jobs to AI. Originality/value – The research provides a theoretical contribution to the growing AI-HRM literature in the context of the theory of cost economics in the context of recruitment efficiency as well as leveraging adaptive learning from the context of the multi-level organizational learning framework to improve the performance of the HR function. The research also provides significant managerial insights for CHROs recommending that they embrace humanized AI in the HRM function and institutionalize AI ethics. © Emerald Publishing Limited}, keywords = {AI ethics; Artificial intelligence; Deskilling; Humanized AI; Human–machine collaboration; Productivity}, } @article{vacekd2026130, title = {Responsible artificial intelligence?}, author = {Vacek D.}, year = {2026}, journal = {AI and Society}, volume = {41}, number = {2}, pages = {843-855}, doi = {10.1007/s00146-025-02604-3}, abstract = {Although the phrase “responsible AI” is widely used in the AI industry, its meaning remains unclear. One can make sense of it indirectly, insofar as various notions of responsibility unproblematically attach to those involved in the creation and operation of AI technologies. It is less clear, however, whether the phrase makes sense when understood directly, that is, as the ascription of some sort of responsibility to AI systems themselves. This paper argues in the affirmative, drawing on a philosophically undemanding notion of role responsibility, and highlights the main consequences of this proposal for AI ethics. © The Author(s) 2025.}, keywords = {Artificial intelligence; Intelligent technologies; Responsible AI; Responsible robotics; Role responsibility; Ethical technology; AI systems; AI Technologies; Intelligent technology; Responsible AI; Responsible robotic; Role responsibility; Intelligent robots}, } @article{arenalaaguadojmarmuacramossfeijooc2026131, title = {Artificial intelligence in the music streaming value chain: Exploring artists' and users’ perceptions on content creation and algorithmic consumption}, author = {Arenal A. and Aguado J.M. and Armuña C. and Ramos S. and Feijoo C.}, year = {2026}, journal = {Telecommunications Policy}, volume = {50}, number = {2}, pages = {103129}, doi = {10.1016/j.telpol.2025.103129}, abstract = {Artificial Intelligence (AI) plays a pivotal role throughout every stage of the music industry value chain. In fact, AI has been integral to the value proposition of music streaming platforms since their inception. This research article investigates how AI—and in particular, Generative AI (GenAI)—affects the creation and consumption stages, the two most socially significant components of the music streaming value chain. It explores users’ perceptions of the impact of AI/GenAI on their music-listening experience and examines how artists and performers view the role and influence of AI on their position and opportunities within the streaming model. Drawing on the findings from two separate focus groups with users and artists/performers from different geographies and professional development, this study complements industry debates—often dominated by technology companies and record and publishing firms—by providing valuable insights into the perceptions at both ends of the music industry value chain regarding the impact of these technologies on music creation, dissemination, and consumption. As key findings, while users exhibited a nuanced response to AI-generated music, both existing literature and insights from artists and performers suggest that AI may further amplify the endemic dysfunctions of music streaming platforms, arising governance issues and ethical concerns, particularly regarding to transparency. In addition, both groups highlighted a significant paradox: while AI has the potential to democratise music creation by lowering barriers to entry, it also poses a threat to the existing ecosystem of music professionals, which have relevant implications in terms of the role of culture in societies, policy and practice. © 2025}, keywords = {AI Ethics and Copyright in Digital Music; AI-Generated Music; Algorithmic Curation and Music Discovery; Artists' perceptions of AI; Generative AI in Music Streaming; Users' perceptions of AI; Computer music; Ethical aspects; Algorithmic curation and music discovery; Algorithmics; Artificial intelligence ethic and copyright in digital music; Artificial intelligence-generated music; Artist' perception of artificial intelligence; Curation; Digital music; Generative artificial intelligence in music streaming; Music streaming; User perceptions; User' perception of artificial intelligence; Acoustic streaming}, } @article{mckennar2026132, title = {Sophistry on steroids? The ethics, epistemology and politics of persuasive AI}, author = {McKenna R.}, year = {2026}, journal = {AI and Society}, volume = {41}, number = {2}, pages = {883-894}, doi = {10.1007/s00146-025-02624-z}, abstract = {This paper examines the ethical, epistemological, and political implications of persuasive AI technologies. Recent research suggests that AI is roughly as persuasive as humans in many contexts. Should this concern us? I argue that, while some worries about persuasive AI may be overblown, we should be worried for a mix of ethical, epistemological and political reasons. Most centrally, we should be worried because persuasive AI may lead to a small number of powerful actors dominating what I call the “marketplace of arguments”—the set of arguments that provide the materials we use to discuss important moral, political and societal issues. © The Author(s) 2025.}, keywords = {AI ethics; influence; marketplace of ideas; persuasion; propaganda; Commerce; Ethical technology; AI ethic; AI Technologies; Ethical implications; Influence; Marketplace of idea; Moral issues; Persuasion; Political implications; Propaganda; Recent researches; Artificial intelligence}, } @article{debbiswasdsenguptar2026133, title = {Reimagining leadership for the AI era: a grounded theory of adaptive, ethical and contextually situated practices in workplaces}, author = {Deb Biswas D. and Sengupta R.}, year = {2026}, journal = {Leadership and Organization Development Journal}, volume = {47}, number = {2}, pages = {395-416}, doi = {10.1108/LODJ-06-2025-0488}, abstract = {Purpose – As artificial intelligence (AI) reshapes workplace systems, traditional leadership models struggle to address the complexities of human-AI collaboration. This study explores the evolving competencies, behaviors, and identity transformations necessary for effective leadership in AI-augmented environments, particularly within the Indian context. Design/methodology/approach – Employing a grounded theory methodology, 60 in-depth interviews were conducted with leaders across diverse industries in India. Data were analyzed inductively to build a contextually rich, systemic framework of leadership in the AI era. Findings – The study identifies a three-pillar model of leadership transformation in AI-augmented workplaces: anticipatory-ethical leadership, socio-technical reengineering, and ontological shifts in leader identity. Leaders proactively address AI's ethical challenges, redesign organizational practices to balance technology and human values, and shift from control-based to distributed, dialogic leadership styles to foster innovation and collective adaptation. Practical implications – Recommendations are offered for HR practitioners and organizational leaders to navigate AI transitions through foresight, empathy, and systemic awareness. Emphasis is placed on fostering distributed leadership, ethical AI integration, and innovation ecosystems that elevate human potential alongside technological advances. Social implications – Beyond organizational practices, the AI-era leadership framework has broad implications for public policy and inter-organizational governance, especially in emerging economies. Leadership development must extend beyond firms into national education systems, ethical AI policies, and multi-stakeholder governance. Categories such as ethical tech sense making and equity tracing can guide certification programs and accountability standards, while inclusive AI socialization and role reengineering inform leadership training in universities and public service academies. At a cross-sector level, shared governance alliances can set norms for refusal criteria, recertification, and equity protections. Ultimately, AI leadership demands systemic, multi-level capacity building for inclusive, sustainable development. Originality/value – The study advances leadership theory by integrating cultural context, emotional labor, and collective resilience into discussions of AI-era leadership. It highlights the systemic, relational, and ethical dimensions that are critical but often overlooked in AI transition narratives. © Emerald Publishing Limited}, keywords = {AI-Augmented leadership; Grounded theory approach; Human-AI collaboration; Leadership identity transformation; Organizational adaptation}, } @article{alamkia2026134, title = {Expanding AI adoption in public sector organizations: perspectives on management practices}, author = {Alamäki A.}, year = {2026}, journal = {Transforming Government: People, Process and Policy}, volume = {20}, number = {2}, pages = {165-187}, doi = {10.1108/TG-05-2025-0124}, abstract = {Purpose – The purpose of this study is to enhance the understanding of management practices that expand artificial intelligence (AI) adoption in public organizations. Design/methodology/approach – The research approach is an exploratory study. Research data were collected from informants representing various organizations working for public or government services in Finland. Findings – The findings of this study indicate that large-scale AI adoption is a complex process in the public sector. This study identified three main practices of AI adoption: technological design practices, AI project design and management practices and networking practices. Additionally, this study shows that several value drivers and barriers to AI adoption are related to technological, organizational and environmental dimensions. This study emphasizes the importance of developing cross-functional AI capabilities and resources, which are crucial for expanding AI initiatives across organizations and networks. Practical implications – Expanding and scaling AI adoption across organizational boundaries requires new management practices and multidisciplinary teamwork, from technological skills to AI governance practices and change management. Originality/value – This study contributes to the emerging research on management practices involved in AI adoption in the public sector. © 2025 Ari Alamäki.}, keywords = {AI adoption; AI governance; Artificial intelligence; Public sector}, } @article{singuhbchakrabortydtroiseccamillerimabrescianis2026135, title = {Responsible AI for trustworthy tourism: A framework for mitigating ambiguity and anxiety with generative AI}, author = {Singu H.B. and Chakraborty D. and Troise C. and Camilleri M.A. and Bresciani S.}, year = {2026}, journal = {Technological Forecasting and Social Change}, volume = {223}, pages = {124407}, doi = {10.1016/j.techfore.2025.124407}, abstract = {Generative AI models are increasingly adopted in tourism marketing content based on text, image, video, and code, which generates new content as per the needs of users. The potential uses of generative AI are promising; nonetheless, it also raises ethical concerns that affect various stakeholders. Therefore, this research, which comprises two experimental studies, aims to investigate the enablers and the inhibitors of generative AI usage. Studies 1 (n = 403 participants) and 2 (n = 379 participants) applied a 2 × 2 between-subjects factorial design in which cognitive load, personalized recommendations, and perceived controllability were independently manipulated. The initial study examined the probability of reducing the cognitive load (reduction/increase) due to the manual search for tourism information. The second study considers the probability of receiving personalized recommendations using generative AI features on tourism websites. Perceived controllability was treated as a moderator in each study. The impact of the cognitive load produced mixed results (i.e., predicting perceived fairness and environmental well-being), with no responsible AI system constructs explaining trust within Study 1. In study 2, personalized recommendations explained each responsible AI system construct, though only perceived fairness and environmental well-being significantly explained trust in generative AI. Perceived controllability was a significant moderator in all relationships within study 2. Hence, to design and execute generative AI systems in the tourism domain, professionals should incorporate ethical concerns and user-empowerment strategies to build trust, thereby supporting the responsible and ethical use of AI that aligns with users and society. From a practical standpoint, the research provides recommendations on increasing user trust through the incorporation of controllability and transparency features in AI-powered platforms within tourism. From a theoretical perspective, it enriches the Technology Threat Avoidance Theory by incorporating ethical design considerations as fundamental factors influencing threat appraisal and trust. © 2025 Elsevier Inc.}, keywords = {Anxiety; Generative AI; Responsible AI; Tourism; Trustworthy; Artificial intelligence; Controllability; Ethical technology; AI systems; Anxiety; Cognitive loads; Ethical concerns; Generative AI; Perceived controllabilities; Personalized recommendation; Responsible AI; Trustworthy; Well being; artificial intelligence; cognition; experimental study; tourism development; World Wide Web; Tourism}, } @article{lislamjfizhanj2026136, title = {A study on the dynamic governance mechanism of digital publishing policies driven by generative AI technology—based on an analytical framework of technological-institutional co-evolution}, author = {Li S. and Lam J.F.I. and Zhan J.}, year = {2026}, journal = {Frontiers in Political Science}, volume = {8}, pages = {1806424}, doi = {10.3389/fpos.2026.1806424}, abstract = {Introduction – The rapid evolution of generative artificial intelligence (GAI) is disrupting the digital publishing sector, creating governance challenges such as ambiguous copyright ownership and unclear platform liability. Existing research often interprets the technology-institution relationship through a unidirectional causal lens, lacking empirical analysis of their interactive mechanisms. This study aims to analyze the co-evolutionary dynamics between GAI and institutional responses to understand how policy systems adapt to technological change. Methods – This study employs a technological-institutional co-evolutionary framework using a mixed-methods approach. The methodology integrates natural language processing (NLP) topic modeling, judicial case coding, and a breakpoint test. The analysis compares 48 policy documents and 14 judicial cases from China, Europe, and the United States, spanning the period from 2016 to 2025. Results – The findings reveal that GAI has driven a structural shift in policy agendas toward AI governance and copyright issues. Comparative analysis shows divergent evolutionary trajectories: China exhibited administration-led catching-up characteristics with a policy lag of approximately 12 months, whereas Europe and the United States demonstrated collaborative adaptation patterns with a longer lag of approximately 24 months. Legal conflicts were predominantly concentrated in the attribution of copyright for AI-generated content (40.63% of cases) and platform liability (35.94%). Discussion – This study reveals the non-linear structural breaks and divergent evolutionary trajectories of institutional responses to GAI. By providing empirical evidence of how different governance systems navigate the balance between technological change and institutional inertia, the findings contribute to the development of adaptive AI governance strategies. Copyright © 2026 Li, Lam and Zhan.}, keywords = {copyright; digital publishing; dynamic governance; generative artificial intelligence; resilient governance; technological-institutional co-evolution}, } @article{coronadoe2026137, title = {From Large Language Models to Agentic AI in Industry 5.0 and the Post-ChatGPT Era: A Socio-Technical Framework and Review on Human–Robot Collaboration}, author = {Coronado E.}, year = {2026}, journal = {Robotics}, volume = {15}, number = {3}, pages = {58}, doi = {10.3390/robotics15030058}, abstract = {Generative Artificial Intelligence (GenAI), particularly Foundation Models (FMs), has recently become a key component of Industry 5.0. Despite growing interest in integrating these technologies into industrial environments, comprehensive analyses of the socio-technical opportunities and challenges of deploying these emerging AI systems in real-world settings remain limited. This article proposes a socio-technical conceptual perspective, termed Responsible Agentic Robotics (RAR), which structures the lifecycle deployment of agentic AI-enabled robotic systems around three core layers: context, design, and value. Additionally, this article presents a brief review of 21 peer-reviewed studies published between 2023 and 2025 (post-ChatGPT era) on FMs and agentic AI-enabled Human–Robot Collaboration (HRC) in industrial assembly/disassembly environments. The results indicate that existing research remains predominantly technology-centric, with a strong emphasis on enhancing robot autonomy, while comparatively limited attention is devoted to human-centered and responsible practices. Moreover, empirical evaluations of human, social, and sustainability dimensions, such as worker empowerment, human factors, well-being, inclusivity, resource utilization, and environmental impact, are rarely conducted and poorly discussed. This article concludes by identifying key socio-technical gaps, outlining future research directions. © 2026 by the author.}, keywords = {agentic AI; Foundation Models; human-centered design; human-robot collaboration; Industry 5.0; large language models; multimodal large language models; responsible AI; socio-technical systems theory; vision-language models}, } @article{huangcwangzwenxwangxyaox2026138, title = {A Conceptual Framework of AI Transparency from Sociotechnical Perspective}, author = {Huang C. and Wang Z. and Wen X. and Wang X. and Yao X.}, year = {2026}, journal = {IEEE Transactions on Artificial Intelligence}, volume = {7}, number = {4}, pages = {1942-1955}, doi = {10.1109/TAI.2025.3608735}, abstract = {Transparency serves as a cornerstone principle in artificial intelligence (AI) ethics and governance, playing a crucial role in upholding ethical standards and ensuring responsible AI deployment. Despite its critical importance, the concept of AI transparency remains fragmented in literature, highlighting the necessity for a unified and comprehensive understanding. This article addresses this imperative by firstly conducting a systematic literature review about existing varied definitions of transparency to deepen our understanding of AI transparency. Then, the three key aspects of AI transparency, that is: 1) transparent to whom; 2) transparent of what; and 3) how to be transparent, are examined. Building upon this groundwork, we propose a novel sociotechnical framework that uniquely integrates both intrinsic and extrinsic dimensions of AI transparency while accounting for the roles of internal and external stakeholders, resulting in a three-layered AI transparency framework encompassing intrinsic, internal, and external transparency. This comprehensive framework not only deepens our understanding of AI transparency but also provides a structured roadmap for navigating the complex sociotechnical landscape of AI systems. Our main contribution lies in developing, this novel conceptual framework that enriches both the theory and practice of AI transparency and provides guidelines for designing and deploying transparent AI systems in the future. © 2020 IEEE.}, keywords = {AI ethics; AI governance; Artificial intelligence (AI); explainable AI; transparency; Artificial intelligence; Ethical technology; Sustainable development; Artificial intelligence ethic; Artificial intelligence governance; Artificial intelligence systems; Conceptual frameworks; Ethical standards; Explainable artificial intelligence; External stakeholders; Socio-technical perspective; Sociotechnical; Systematic literature review; Transparency}, } @article{desaiddesaia2026139, title = {From Models to Metrics: A Governance Framework for Large Language Models in Enterprise AI and Analytics}, author = {Desai D. and Desai A.}, year = {2026}, journal = {Analytics}, volume = {5}, number = {1}, pages = {8}, doi = {10.3390/analytics5010008}, abstract = {Large language models (LLMs) and other foundation models are rapidly being woven into enterprise analytics workflows, where they assist with data exploration, forecasting, decision support, and automation. These systems can feel like powerful new teammates: creative, scalable, and tireless. Yet they also introduce distinctive risks related to opacity, brittleness, bias, and misalignment with organizational goals. Existing work on AI ethics, alignment, and governance provides valuable principles and technical safeguards, but enterprises still lack practical frameworks that connect these ideas to the specific metrics, controls, and workflows by which analytics teams design, deploy, and monitor LLM-powered systems. This paper proposes a conceptual governance framework for enterprise AI and analytics that is explicitly centered on LLMs embedded in analytics pipelines. The framework adopts a three-layered perspective—model and data alignment, system and workflow alignment, and ecosystem and governance alignment—that links technical properties of models to enterprise analytics practices, performance indicators, and oversight mechanisms. In practical terms, the framework shows how model and workflow choices translate into concrete metrics and inform real deployment, monitoring, and scaling decisions for LLM-powered analytics. We also illustrate how this framework can guide the design of controls for metrics, monitoring, human-in-the-loop structures, and incident response in LLM-driven analytics. The paper concludes with implications for analytics leaders and governance teams seeking to operationalize responsible, scalable use of LLMs in enterprise settings. © 2026 by the authors.}, keywords = {AI governance; algorithmic auditing; enterprise AI; enterprise analytics; large language models; responsible AI}, } @article{lambertjmlaskovaiaagaraninaobogatyrevak2026140, title = {Smarter, not harder: the AI capability paradox in emerging-market SMEs}, author = {Lambert J.M. and Laskovaia A. and Garanina O. and Bogatyreva K.}, year = {2026}, journal = {Journal of Entrepreneurship in Emerging Economies}, volume = {18}, number = {3}, pages = {813-836}, doi = {10.1108/JEEE-10-2025-0632}, abstract = {Purpose – This study aims to identify configurations of artificial intelligence (AI)-related organisational capabilities that lead to superior performance in small and medium-sized enterprises (SMEs) operating in an emerging market, moving beyond the assumption that “more AI usage is better”. Drawing on resource orchestration theory, the authors conceptualise how governance, skills, ethics, leadership and data infrastructure jointly enable value creation from AI. Design/methodology/approach – This study relies on a cross-sectional survey of Russian SMEs (October 2024 to January 2025). Of 384 firms, 47 that reported AI use were analysed. Using fuzzy-set qualitative comparative analysis (fsQCA), the authors examined how AI usage intensity combines with internal enablers, AI governance, data infrastructure, employee AI/digital skills, top management team (TMT) involvement and AI ethics preparedness, to explain four outcomes: operational efficiency, strategic decision quality, product/service innovation and customer responsiveness. The authors calibrated the conditions using the direct method and explored the robustness of configurations across alternative consistency and frequency thresholds. Findings – Across all outcomes, high AI usage intensity was not a core condition. Instead, multiple high-performance pathways featured AI governance as a central ingredient, frequently complemented by ethics preparedness and either employee training or active TMT involvement. Where governance was weaker, strong employee capabilities could serve as a substitute. These results show that SMEs can achieve strong performance with moderate AI intensity when organisational capabilities are well-aligned. In emerging-market SMEs, this points to an “AI capability paradox”: under the dual constraints of limited resources and weaker institutional environments, more intensive AI use does not necessarily yield better outcomes unless complemented by appropriate capability bundles. Originality/value – The authors shift the debate from “how much AI” to “how AI is governed and supported”. By applying a configurational lens in an emerging-market SME context, the authors reveal equifinal capability bundles, highlighting governance and ethics, paired with skills and leadership, as more decisive than sheer adoption intensity. The authors extend AI-related capabilities research to the focus on SMEs in emerging markets. Methodologically, the authors use fsQCA to identify multiple, empirically grounded resource and capability configurations associated with superior performance. © 2026 Emerald Publishing Limited}, keywords = {Artificial intelligence; Digital technology; fsQCA; Management of innovation; SMEs}, } @article{pigact2026141, title = {Transparency in large language model (LLM)-powered digital human twins: the AI ethics perspective}, author = {Pigac T.}, year = {2026}, journal = {AI and Society}, volume = {41}, number = {3}, pages = {2109-2118}, doi = {10.1007/s00146-025-02617-y}, abstract = {Digital human twins (DHTs), powered by large language models (LLMs), are transforming industries such as healthcare and finance by mimicking human behaviors, preferences, and decision-making processes. While their adoption offers unprecedented personalization and engagement, it also raises significant ethical concerns, particularly regarding transparency. Ensuring users understand how these systems function is critical to fostering trust and accountability. This study explores transparency in LLM-powered DHTs through qualitative analysis of 30 semi-structured interviews with users across diverse sectors. The findings reveal critical challenges, including algorithmic opacity, data privacy vulnerabilities, and threats to user autonomy. Participants consistently expressed a need for clear disclosures about data practices and emphasized the importance of robust ethical safeguards to prevent misuse. The research highlights the tension between achieving transparency and maintaining the seamless functionality of DHT systems. It underscores the risks of oversimplifying algorithmic processes while pointing out the erosion of trust caused by opaque operations. To address these challenges, the study proposes actionable strategies, including tiered transparency models, enhanced regulatory oversight, and user-centric design principles. By bridging ethical principles with practical applications, this research provides a roadmap for fostering responsible AI innovation. It advances the discourse on ethical AI by addressing transparency challenges in LLM-powered DHTs, emphasizing the need for systems that uphold trust, accountability, and user autonomy. © The Author(s), under exclusive licence to Springer-Verlag London Ltd., part of Springer Nature 2025.}, keywords = {AI ethics; Data privacy; Digital human twins; LLM; Personalization; Transparency; Behavioral research; Decision making; Ethical technology; AI ethic; Decision-making process; Digital human twin; Digital humans; Human behaviors; Human decisions; Language model; Large language model; Personalizations; User autonomy; Data privacy; Transparency}, } @article{limatglpassoswl2026142, title = {When decentralization is not enough: Federated Learning and the institutional, infrastructural, and regulatory conditions for AI governance in Brazil}, author = {Lima T.G.L. and Passos W.L.}, year = {2026}, journal = {Computer Law and Security Review}, volume = {60}, pages = {106277}, doi = {10.1016/j.clsr.2026.106277}, abstract = {Federated Learning (FL) is often framed as a privacy-oriented alternative to centralized training, but decentralizing optimization does not necessarily decentralize Artificial Intelligence (AI) governance. Using Brazil as the case, we ask when FL is a feasible alternative — given uneven public-sector digital capacity and dependence on external cloud/software stacks, rather than a technical fix. We ask under which infrastructural, technical, and institutional constraints FL could reduce record-level data transfers without re-centralizing control through aggregators, orchestration layers, and evaluation protocols. Bridging Science and Technology Studies (STS), law, and computer engineering, we treat FL as a sociotechnical infrastructure and evaluate it with three benchmarks: (i) enforceable auditability of updates, aggregation, and validation; (ii) participatory governance (contestability) over objectives, participation rules, and metrics; and (iii) equitable computational capacity to prevent exclusion of under-resourced institutions. We define audit-ready evidence in federated pipelines and introduce Accountable Federated FactSheets (AF2) to document update provenance, aggregation decisions, and verification claims. We then assess Brazil's governance instruments – EBIA, LGPD, and Draft Bill 2338/2023 – and argue that FL's governance relevance is contingent on enforceable audit hooks, role allocation, and capacity-building, not AI model architecture alone. © 2026 Elsevier Ltd}, keywords = {AI governance; Brazil AI policy; Decentralized machine learning; Federated learning; Artificial intelligence; Public policy; Artificial intelligence governance; Brazil artificial intelligence policy; Centralised; Decentralisation; Decentralised; Decentralized machine learning; Feasible alternatives; Machine-learning; Optimisations; Regulatory conditions; Learning systems}, } @article{cookjrotenbergs2026143, title = {AI innovations, applications and recommendations in property management}, author = {Cook J. and Rotenberg S.}, year = {2026}, journal = {Property Management}, volume = {44}, number = {2}, pages = {154-173}, doi = {10.1108/PM-01-2025-0006}, abstract = {Purpose – The study's central question is how can AI enhance property management (PM) while balancing equity, ethics and sustainability? This question reveals opportunities and challenges. While AI fosters innovation and streamlines operations, it raises ethical concerns such as data privacy, algorithmic biases and risks of marginalizing vulnerable populations. Design/methodology/approach – A literature review was conducted on the current research on AI in property management. Afterward, the research focused on real estate investment, predictive maintenance and the tenant experience. This information was then used to make recommendations for property managers interested in AI services. Findings – The findings indicate that AI has the potential to significantly enhance various aspects of property management as long as there is human oversight. Research limitations/implications – While AI offers significant potential, challenges such as data quality, algorithmic bias and cybersecurity risks must be addressed. Continued research is needed to fully understand the ethical implications of AI in property management, as well as to develop a framework for using AI responsibly. Practical implications – The paper demonstrates that AI can streamline core property management functions by enhancing predictive maintenance, improving valuation accuracy, and automating routine tasks such as rent collection and service requests. These tools help managers reduce operational costs, increase efficiency, and make more informed decisions through data-driven insights. When paired with strong human oversight, AI also strengthens security and sustainability practices, offering managers a practical pathway to modernize daily operations. Social implications – Socially, the paper highlights how AI can enhance tenant experiences through faster communication, personalized services, and safer living environments. However, it also warns of risks such as algorithmic bias, unequal access to technology, and data privacy concerns that may disproportionately affect vulnerable populations. The research emphasizes the need for ethical guidelines and transparent governance to ensure AI supports fairness, inclusivity, and trust within communities. Originality/value – This research paper contributes to the growing but relatively new field of AI in property management. It provides an overview of AI use in property management and identifies areas for future research. © Emerald Publishing Limited}, keywords = {Artificial intelligence (AI); Machine learning; Predictive maintenance; Property management; Real estate investment; Tenant experience}, } @article{buijsmans2026144, title = {Accuracy is not all you need! The Reasons to Require AI Explainability}, author = {Buijsman S.}, year = {2026}, journal = {Minds and Machines}, volume = {36}, number = {1}, pages = {14}, doi = {10.1007/s11023-026-09768-x}, abstract = {Do we need explanations of AI outputs in order to use AI systems (in high-risk settings)? This question has been actively debated recently, with one group denying that explanations are needed as long as the AI system is sufficiently accurate. What matters, according to them, is that outcomes improve. The other group argues that we have procedural reasons, centered around autonomy and self-advocacy, which trump outcome-based arguments to the contrary. I here present a set of arguments to show that outcome-based arguments should in fact also favor explainability for many of the current systems, as challenges with human oversight and accountability often lead to worse overall outcomes even if a more accurate AI system is integrated. Critics of explainability overlooked the fact that AI operates within a broader socio-technical system, and its accuracy alone tells us little of the final outcomes. In addition, I consolidate the procedural arguments and present a view of the upshot of these arguments. On this, we should avoid applications of AI that largely replace decision-making (relegating humans to the position of checking outputs). We can, however, use AI in other roles even for high-risk decision making while conforming to all of the requirements set by both outcome-based and procedural arguments. What matters, in the end, is the ability to explain decisions, and with the right role for AI that is possible even when supported by opaque systems. © The Author(s) 2026.}, keywords = {AI Ethics; Explainable AI; Explanation; Procedural justice; Right to explanation; Artificial intelligence; AI ethic; AI systems; Current system; Explainable AI; Explanation; Human oversight; Procedural justice; Right to explanation; Self-advocacy; Sociotechnical systems; Decision making}, } @article{takemia2026145, title = {Bridge-weaving diplomacy: Japan’s strategy for AI governance in navigating geopolitical fragmentation}, author = {Takemi A.}, year = {2026}, journal = {Journal of Digital Media and Policy}, volume = {17}, number = {1}, pages = {151-170}, doi = {10.1386/jdmp_00203_1}, abstract = {Global AI governance is taking on geopolitical dimensions as major powers pursue divergent regulatory models. In this context of normative fragmentation – visible both in intra-G7 divergences and in potential tensions between G7-led frameworks and sovereignty-emphasizing approaches advanced by other major actors – this article analyses Japan’s diplomatic role. The article conceptualizes Japan’s strategy as ‘bridge-weaving diplomacy’ and examines its two-stage process. First, it ‘bridged’ divergent approaches within the G7 to forge an interoperable consensus – the Hiroshima AI Process (HAIP) – in cooperation with other member states. Second, it has sought to ‘weave’ this consensus beyond the G7 into a more inclusive framework, notably through the establishment of the HAIP Friends Group, as well as its cooperative postures towards other nations including ASEAN countries. Drawing on an analysis of events from 2023 to 2025, official sources and interviews with key informants, the article argues that this approach can be taken as a critical enabling role in mitigating normative fragmentation while enhancing the legitimacy of a governance model grounded in liberal-democratic principles (human rights, rule of law and multi-stakeholder participation), amidst potential structural tensions with state-sovereignty-oriented approaches. Japan’s case provides a granular analysis of middle-power statecraft amid geopolitical competition, offering a new perspective on debates surrounding norm formation. © 2026 Intellect Ltd.}, keywords = {digital sovereignty; emerging technology; geopolitical competition; global governance; HAIP; interoperability; middle-power diplomacy}, } @article{dwivediykhelalmyielgendyiaalahmadrwaltonpsuhasinghvjeoni2026146, title = {Agentic AI Systems: What It Is and Isn't}, author = {Dwivedi Y.K. and Helal M.Y.I. and Elgendy I.A. and Alahmad R. and Walton P. and Suh A. and Singh V. and Jeon I.}, year = {2026}, journal = {Global Business and Organizational Excellence}, volume = {45}, number = {3}, pages = {253-263}, doi = {10.1002/joe.70018}, abstract = {The rapid adoption of artificial intelligence (AI) is shifting from tools that assist human tasks toward self-directed, agentic AI systems capable of planning and executing complex goals with minimal oversight. However, a clear understanding of what distinguishes these systems from conventional AI agents and generative AI is lacking, obscuring their unique opportunities and risks. To this end, this article addresses that gap by defining the core concepts, technologies, and management approaches for agentic AI systems, which utilize planning, shared memory, tools, and multi-agent teamwork to complete complex tasks autonomously. By contrasting this paradigm with its predecessors, the paper synthesizes recent technical surveys, governance proposals, and early industrial deployments to highlight that while agentic AI enables transformative applications like end-to-end process automation and adaptive decision support, it also introduces significant challenges, including cascading errors, goal misalignment, and regulatory gaps. Finally, this paper concludes with strategic guidance for organizations and consumers to adopt the capabilities of these systems responsibly, emphasizing the imperative of maintaining transparency, accountability, and human oversight. © 2025 Wiley Periodicals LLC.}, keywords = {adaptive decision-making; agentic AI systems; AI agents; AI autonomy; AI governance frameworks; business process automation; generative AI; multi-agent orchestration; persistent memory architectures; planning loops; tool-augmented reasoning}, } @article{daisliqjiasjliugkinclthajlin2026147, title = {Responsible AI in knowledge creation: An exploration of generative AI's opportunities and risks}, author = {Dai S. and Li Q. and Jia S.J. and Liu G. and Kincl T. and Hajli N.}, year = {2026}, journal = {Technological Forecasting and Social Change}, volume = {226}, pages = {124570}, doi = {10.1016/j.techfore.2026.124570}, abstract = {This study explores the transformative potential and inherent challenges of Generative AI in the domain of knowledge creation and management, using the Socialization, Externalization, Combination, and Internalization (SECI) model as an analytical framework. Our qualitative research, based on content analysis from expert opinions, reveals that the integration of Generative AI in knowledge processes is inevitable and offers substantial productivity enhancements. These include providing diverse expression channels, simulating personalized interactions, and facilitating cross-disciplinary communication. However, significant risks accompany these benefits, such as threats to data security, personal privacy, and intellectual property, as well as issues of misinformation, data bias, and reduced human cognitive engagement. The findings extend the SECI model by highlighting specific challenges posed by AI technologies at each knowledge creation stage: socialization, externalization, combination, and internalization. The study underscores the necessity of a balanced approach, integrating technological, ethical, and socio-cultural perspectives to evaluate AI's impact comprehensively. Our research contributes to the theoretical understanding of AI's role in knowledge management and offers actionable strategies for its ethical and effective implementation, emphasizing the importance of interdisciplinary approaches and continuous regulatory adaptation. © 2026 Elsevier Inc.}, keywords = {AI risks; Generative AI; Knowledge creation; Knowledge management; Technological advancements; Data privacy; Domain Knowledge; Ethical technology; Knowledge acquisition; Knowledge transfer; AI risk; Content analysis; Domain of knowledge; Expert opinion; Generative AI; Knowledge creations; Knowledge process; Productivity enhancement; Qualitative research; Technological advancement; artificial intelligence; cognition; interdisciplinary approach; knowledge; risk assessment; technological development; Knowledge management}, } @article{subasbeltrnpuncetaidelecuonaipujolo2026148, title = {Asking the right questions: a governance approach to uphold human autonomy in artificial intelligence}, author = {Subías-Beltrán P. and Unceta I. and de Lecuona I. and Pujol O.}, year = {2026}, journal = {AI and Society}, volume = {41}, number = {2}, pages = {1149-1173}, doi = {10.1007/s00146-025-02611-4}, abstract = {This paper explores the impacts of artificial intelligence (AI) systems on human autonomy and identifies the responsibilities of different stakeholders in addressing them. As AI systems increasingly shape not only the outcomes but also the conditions of everyday decision making, they directly influence individuals’ ability to exercise free and informed choice, a capacity central to democratic participation. Hence, it becomes crucial to examine how the definition, design, and implementation of AI systems may effectively support or constrain autonomy, through a well-defined organizational governance framework. Yet, most existing governance approaches overlook or marginalize the role of autonomy, leaving a critical gap. To address this, we introduce a governance framework structured around a set of diagnostic questions that link design choices to concrete dimensions of human autonomy at both system and user levels, and that map these impacts across the different stages of the decision-making process. The framework clarifies how responsibility for answering these questions can be distributed across different organizational roles, promoting internal reflection and collaboration. A dedicated section illustrates how the framework can be applied in practice to recognize and address autonomy impacts in AI systems within different organizational settings and development processes. This paper provides a practical tool to support autonomy-centred AI governance, grounded in the view that upholding human autonomy today is essential to secure sustainable and democratic futures tomorrow. © The Author(s) 2025.}, keywords = {AI governance; Artificial intelligence; Human autonomy; Value sensitive design; Decision making; Artificial intelligence governance; Artificial intelligence systems; Condition; Decisions makings; Democratic participation; Design and implementations; Human autonomy; Link design; Organisational; Value sensitive design; Artificial intelligence}, } @article{mabensklambiasejvasquezra2026149, title = {Self-starters and lone rangers: Municipal government and nonprofit PR practitioners’ approaches to AI training, ethics, and policy-making}, author = {Maben S.K. and Lambiase J. and Vasquez R.A.}, year = {2026}, journal = {Public Relations Review}, volume = {52}, number = {1}, pages = {102664}, doi = {10.1016/j.pubrev.2025.102664}, abstract = {Researchers surveyed municipal government and nonprofit communicators (n = 195) to investigate their approaches to generative AI technologies in regards to ethics. The study aimed to discover challenges related to moral silence and how practitioners explored technology and guidelines for its use. Municipal government and nonprofit communicators reported on their AI adoption plans, usage, and ethical foundations. A nine-question Ethical Sensitivity Scale (ESS) was adapted for use (α =.85) in the survey, and the sample represented an ethically sensitive group ( M = 4.32 on a 5-point scale). Findings indicate that 85 % of the communicators were using generative AI tools in their work as professional communicators, but were doing so with little or no training, planning, internal discussions, or ethical guidelines. Respondents’ highest concerns about ethics and generative AI (intellectual property, copyright, plagiarism, honesty, and accuracy) did not match the concerns they perceived as most important to their supervisors. A fifth even suggested that their supervisors had no ethical concerns related to AI. Practical and theoretical implications are explored, and a new model for ethical adoption of technology in public relations practice is proposed. © 2025 Elsevier Inc.}, keywords = {Ethical silence; Ethics; Generative artificial intelligence; Moral muteness; Moral myopia; Municipal government; Nonprofit}, } @article{hardcastlefhennekharbjileeavianajnhalfords2026150, title = {Unblackboxing How Sociotemporalities Inform AI Accountability: The Case of Targeted Advertising}, author = {Hardcastle F. and Henne K. and Harb J.I. and Lee A. and Viana J.N. and Halford S.}, year = {2026}, journal = {Social Science Computer Review}, volume = {44}, number = {1}, pages = {131-149}, doi = {10.1177/08944393251365275}, abstract = {In recent years, numerous accountability interventions have been introduced to address the harms and inequalities associated with Artificial Intelligence (AI) systems. Early efforts concentrated on transparency and explainability, often operationalized as technical fixes intended to “open the black box” and render algorithmic processes more intelligible. However, sociological research has revealed the limitations of these interventions, particularly their narrow focus on the technical and operational dimensions of AI. In response, sociologists have broadened the scope of “unblackboxing” to include the sociomaterialities of AI, or the social, political, and environmental relations that both shape and are shaped by AI technologies and their infrastructures. This article extends this agenda by focusing on sociotemporalities: the narratives and structures of time that shape both AI systems and the interventions meant to improve their accountability. We re-analyze interviews from a retrospective study of transparency in targeted advertising, a domain long associated with concerns about privacy, discrimination and opaque practices. Drawing on the sociology of time, and especially the sociology of the future, we examine the sociotemporalities that actively shaped the development of targeted advertising technologies at the time and influenced key informants’ thinking about approaches to improve accountability. Our analysis suggests that sociotemporalities exerted a structuring influence on how “appropriate” accountability interventions were imagined and enacted, ultimately shaping the emergent present of targeted advertising. We discuss the application of such an approach in the context of emerging AI technologies and AI accountability interventions. We conclude by arguing that expanding unblackboxing to include sociotemporal as well as sociomaterial dimensions can help open new pathways for designing and implementing more practical, effective, and context-specific AI accountability interventions. © The Author(s) 2025. This article is distributed under the terms of the Creative Commons Attribution 4.0 License (https://creativecommons.org/licenses/by/4.0/) which permits any use, reproduction and distribution of the work without further permission provided the original work is attributed as specified on the SAGE and Open Access page (https://us.sagepub.com/en-us/nam/open-access-at-sage).}, keywords = {targeted advertising, artificial intelligence, sociotemporalities, sociology of futures, transparency, accountability, timing, pacing}, } @article{seplvedadpmachucarc2026151, title = {Explainable artificial intelligence as a basis for its governance: Implications for civil liability; [La inteligencia artificial explicable como fundamento para su gobernanza: Implicancias en la responsabilidad civil]}, author = {Sepúlveda D.P. and Machuca R.C.}, year = {2026}, journal = {Revista Chilena de Derecho y Tecnologia}, volume = {14}, pages = {e74040}, doi = {10.5354/0719-2584.2025.74317}, abstract = {This article analyzes explainability, firstly, as a requirement for reliability, an ethical principle that serves as a foundation for the governance of artificial intelligence. From this perspective, the concepts of governance, reliability and explainability of artificial intelligence (AI) are analyzed, in line with their evolution, mainly in the framework of the European Union. Secondly, we study whether or not explainable artificial intelligence is relevant to liability regimes for damages caused by AI systems. This allows us to reflect on the importance of incorporating explainability in the new liability schemes that will be generated to address the problems generated in terms of damages caused by AI. © (2026), (Universidad de Chile). All rights reserved.}, keywords = {advanced artificial intelligence; AI governance; civil liability; desarrollo tecnológico; Explainable artificial intelligence; gobernanza de la IA; inteligencia artificial avanzada; Inteligencia artificial explicable; responsabilidad civil; technological development}, } @article{bargilo2026152, title = {Examining trends in AI ethics across countries and institutions via quantitative discourse analysis}, author = {Bar-Gil O.}, year = {2026}, journal = {AI and Society}, volume = {41}, number = {4}, pages = {3297-3312}, doi = {10.1007/s00146-025-02673-4}, abstract = {This study examines how institutional contexts influence the AI ethics landscape through quantified qualitative discourse analysis. We analyzed ten foundational AI ethics documents from academic, industry, military/defense, and national sectors (2018–2021) to investigate whether purportedly universal ethical principles maintain consistent meanings across contexts. The methodology integrated computational frequency analysis of purposive sample, targeting influential texts functioning as obligatory passage points in AI ethics discourse. We identified 14 ethical principles through systematic word list development, analyzed 2351 coded segments across documents, and mapped semantic co-occurrence patterns. The analysis revealed that universal principles undergo systematic recontextualization through institutional appropriation. Privacy transforms from rights-based frameworks in EU documents to security-balance approaches in US military contexts to collective security conceptualizations in Israeli frameworks. Military frameworks uniquely emphasize governability and traceability, while these principles remain absent in academic and industry documents. Industry texts prioritize technical operationalization over distributive justice concerns, with equitability completely absent despite appearing in 15.8% of academic codes. These findings challenge assumptions about universal AI ethics, demonstrating that institutional logics constitute rather than merely influence ethical discourse. Effective AI governance requires acknowledging this contextual heterogeneity through sector-specific guidelines that recognize interpretive variations while maintaining core principles. The research contributes methodologically by demonstrating how quantified qualitative analysis reveals systematic patterns invisible to purely qualitative approaches, and theoretically by establishing institutional positioning as constitutive of ethical meaning in AI governance discourse. © The Author(s) 2025.}, keywords = {AI ethics; AI governance; Code of ethics; Quantitative discourse analysis; Responsible innovation; Artificial intelligence; Ethical technology; Academic industry; AI ethic; AI governance; Code of Ethics; Discourse analysis; Ethical principles; Institutional contexts; Military defense; Quantitative discourse analyze; Responsible innovation; Semantics}, } @article{xingyheyzengl2026153, title = {Unveiling big data-driven price discrimination in China: Evidence from public opinion on Chinese Tiktok}, author = {Xing Y. and He Y. and Zeng L.}, year = {2026}, journal = {Technological Forecasting and Social Change}, volume = {223}, pages = {124445}, doi = {10.1016/j.techfore.2025.124445}, abstract = {The application of advanced artificial intelligence (AI) algorithms in China's big data-driven price discrimination (BDPD) has garnered considerable attention due to its significant impact on consumer welfare and market efficiency. These algorithms enable dynamic pricing adjustments, yet their lack of transparency often leads to public perceptions of inequity and exploitation. This paper presents a case study of Douyin, a prominent Chinese social media platform, to examine public opinion regarding BDPD. Leveraging established AI ethics frameworks, we conduct an inductive qualitative content analysis to identify the ethical principles pertinent to various facets of BDPD. A taxonomy is developed based on four primary AI biases expressed by the Chinese public regarding BDPD: (1) Price increases following search experiences; (2) Price variations across different devices; (3) Discriminatory pricing based on customers' consumption levels; and (4) Customer tagging for discriminatory information delivery. Furthermore, we construct a theoretical framework that informs future research and provide actionable insights for scholars. The results support policy development aimed at addressing public apprehensions about the ethical ramifications of emerging technologies and algorithmic systems, thereby promoting mutually beneficial outcomes. © 2025 Elsevier Inc.}, keywords = {AI bias; Artificial intelligence; Big data; Case study; Price discrimination; Social media; China; Costs; Discriminators; Ethical technology; Social aspects; Social networking (online); Artificial intelligence algorithms; Artificial intelligence bias; Case-studies; Consumer market; Consumer welfares; Data driven; Market efficiency; Price discrimination; Public opinions; Social media; algorithm; artificial intelligence; data management; public attitude; social media; Artificial intelligence; Big data}, } @article{snchezrl2026154, title = {Artificial Intelligence in the Administration of Justice: An Opportunity for Mexico from a Comparative Analysis; [Inteligencia artificial en la impartición de justicia: una oportunidad para México desde un análisis comparado]}, author = {Sánchez R.L.}, year = {2026}, journal = {Nuevo Derecho}, volume = {22}, number = {38}, pages = {1-28}, doi = {10.25057/2500672X.1837}, abstract = {A critical perspective is presented on the potential adoption and implementation of artificial intelligence (AI) in the administration of justice in Mexico. Ten key functions are systematized (electronic case processing, caseflow management, semantic search, classification and clustering, assisted drafting, transcription, translation/plain language, ODR, analytics, and citizen assistants), and international cases are examined, highlighting the institutional factors behind their success and their effects on efficiency and effectiveness. A restrictive notion of AI in justice is adopted; it is conceived as a support tool subject to human oversight. The proposed adoption of AI mechanisms must at all times be guided by the principles of human supervision, transparency, traceability, reasonable explainability, non-discrimination, and information security. The study shows that the best results come from interoperable platforms and model governance, rather than from replacing the judiciary. Atribución – No comercial – Compartir igual 4.0 Internacional. Más información: https://creativecommons.org/licenses/by-nc-nd/4.0/}, keywords = {acceso a la justicia; Access to justice; AI ethics; analítica judicial; artificial intelligence; búsqueda jurídica; case management; digital evidence; evidencia digital; función jurisdiccional; gestión de casos; inteligencia artificial; judicial analytics; judicial function; legal search; ODR; ODR; ética de la IA}, } @article{zhangzlinmliux2026155, title = {Enhancing employees’ change readiness for artificial intelligence (AI) adoption: The role of AI transparency}, author = {Zhang Z. and Lin M. and Liu X.}, year = {2026}, journal = {International Journal of Hospitality Management}, volume = {134}, pages = {104620}, doi = {10.1016/j.ijhm.2026.104620}, abstract = {Based on the heuristic-systematic model (HSM), this study examines how AI transparency enhances employees’ change readiness for AI adoption. Through three scenario-based experiments, this research demonstrates that AI transparency not only directly increases change readiness but also operates through two distinct psychological pathways: improving AI job role clarity and reducing discomfort. Importantly, these indirect effects are observed only under low levels of leaders’ AI symbolization. These findings offer theoretical contributions by elucidating the dual-process mechanisms underlying AI acceptance in human resource management, extend the application of HSM, and provide actionable guidance for leveraging transparency and leadership cues to improve implementation outcomes. © 2026 Elsevier Ltd}, keywords = {AI job role clarity; AI transparency; Discomfort; Employees’ change readiness for AI adoption; Leaders’ AI symbolization}, } @article{phamqtttranhtnguyenvp2026156, title = {Semiotics of algorithmic law in Vietnam: Decoding the digital transformation of legal texts and practices}, author = {Pham Q.T.T. and Tran H.T. and Nguyen V.P.}, year = {2026}, journal = {AI and Society}, volume = {41}, number = {3}, pages = {2487-2510}, doi = {10.1007/s00146-025-02440-5}, abstract = {This study analyzes how algorithmic governance transforms the legal system in Vietnam through its National Program for Digital Transformation (PTDN), thus filling the lacuna in legal semiotics concerning changes from human-centered signification to algorithmic signification. Using a qualitative-interpretive approach, semiotic analysis is applied to urban case studies, weaving the data from e-contracts, AI-constructed administrative decisions, and legal practices (March 2024–February 2025). Findings reveal that algorithms make legal texts computational signifiers, thereby improving efficiency but obscuring intent; favor data over narrative coherence; and supplant human discretion with algorithmic authority, which is very much a factor of Vietnam’s collectivist legal culture. These transformations threaten the substance of justice, and in that regard, transparent algorithmic governance is a must. The placement of the computational interpretant in theoretical development connects semiotic theory with a non-Western perspective on AI law. It thereby equips policymakers to devise accountable regulations, judges to employ a semiotic toolkit for overseeing AI-imposed rulings, and legal practitioners to assist in disbursing efficiency and oversight. This state-led form of digitalization makes clear Vietnam’s somewhat distinct position in world AI governance, therefore opening pathways for longitudinal- and rural-focused studies on AI’s societal implications. © The Author(s), under exclusive licence to Springer-Verlag London Ltd., part of Springer Nature 2025.}, keywords = {AI ethics; Algorithmic law; Computational interpretant; Legal semiotics; State-centric governance; Vietnam digitalization; Computation theory; Ethical aspects; Public policy; Semiotics; AI ethic; Algorithmic law; Algorithmics; Computational interpretant; Digital transformation; Legal semiotic; Legal texts; State-centric governance; Viet Nam; Vietnam digitalization; Efficiency}, } @article{deoliveiracgbalbuquerqueodpbelottiellopesifsilvarbdaarbixg2026157, title = {An analysis of facial recognition in banking disputes through data scraping from São Paulo court of appeal decisions}, author = {de Oliveira C.G.B. and Albuquerque O.D.P. and Belotti E.L. and Lopes I.F. and Silva R.B.D.A. and Arbix G.}, year = {2026}, journal = {Applied Soft Computing}, volume = {189}, pages = {114500}, doi = {10.1016/j.asoc.2025.114500}, abstract = {Context: The increasing adoption of Artificial Intelligence (AI) systems in financial services has led to a growing number of legal disputes concerning the use of facial recognition technology (FRT) in Brazil. This study examines how the São Paulo Court of Appeal has addressed consumer claims related to contracting payroll loans using FRT-based authentication. To achieve this, we employed web scraping techniques to collect 186 judicial decisions referencing AI-related terms and 10,745 decisions specifically mentioning “facial biometrics” from the Court's online portal between 2012 and 2024. The data was processed and filtered using data mining techniques to isolate cases directly involving banking disputes over facial recognition in credit contracts. A qualitative analysis focused on 41 decisions under the AI term search and 418 under the facial biometrics search. The findings reveal a consistent judicial tendency to validate contracts based on FRT authentication, often shifting the burden of proof to consumers, even in cases involving vulnerable individuals such as the elderly. Moreover, courts frequently accept minimal technical evidence from banks regarding the reliability and error rates of their facial recognition systems. These judicial decisions raise concerns about consumer rights protection, informed consent, and judicial reliance on technology perceived as infallible. The paper highlights the need for regulatory oversight by the Brazilian Central Bank and the National Council of Justice to establish technical standards and procedural safeguards when financial institutions use FRT in consumer credit contracts. These findings contribute to ongoing discussions on AI accountability and the ethical deployment of biometric technologies in the financial sector. © 2025 Elsevier B.V.}, keywords = {Artificial intelligence; Facial recognition systems; Jurisprudence; Law; Legal cases; Authentication; Banking; Biometrics; Consumer protection; Data mining; Ethical aspects; Face recognition; Wages; Artificial intelligence systems; Facial recognition; Facial recognition systems; Financial service; Jurisprudence; Law; Legal case; Legal disputes; Sao Paulo; Technology-based; Artificial intelligence}, } @article{laoyyouy2026158, title = {Unraveling generative AI in BBC News: application, impact, literacy and governance}, author = {Lao Y. and You Y.}, year = {2026}, journal = {Transforming Government: People, Process and Policy}, volume = {20}, number = {1}, pages = {28-53}, doi = {10.1108/TG-01-2024-0022}, abstract = {Purpose – This study aims to uncover the ongoing discourse on generative artificial intelligence (AI), literacy and governance while providing nuanced perspectives on stakeholder involvement and recommendations for the effective regulation and utilization of generative AI technologies. Design/methodology/approach – This study chooses generative AI-related online news coverage on BBC News as the case study. Oriented by a case study methodology, this study conducts a qualitative content analysis on 78 news articles related to generative AI. Findings – By analyzing 78 news articles, generative AI is found to be portrayed in the news in the following ways: Generative AI is primarily used in generating texts, images, audio and videos. Generative AI can have both positive and negative impacts on people’s everyday lives. People’s generative AI literacy includes understanding, using and evaluating generative AI and combating generative AI harms. Various stakeholders, encompassing government authorities, industry, organizations/institutions, academia and affected individuals/users, engage in the practice of AI governance concerning generative AI. Originality/value – Based on the findings, this study constructs a framework of competencies and considerations constituting generative AI literacy. Furthermore, this study underscores the role played by government authorities as coordinators who conduct co-governance with other stakeholders regarding generative AI literacy and who possess the legislative authority to offer robust legal safeguards to protect against harm. © 2024 Yucong Lao and Yukun You.}, keywords = {AI governance; AI literacy; BBC News; Generative AI}, } @article{sonsh2026159, title = {Abductive Discretization and Residual Politics: From Kantian Schematism to “Open Schema” AI Governance}, author = {Son S.H.}, year = {2026}, journal = {Philosophies}, volume = {11}, number = {2}, pages = {51}, doi = {10.3390/philosophies11020051}, abstract = {Fairness and minority exclusion have emerged as the central concerns of contemporary Artificial Intelligence (AI) ethics. However, standard auditing and documentation practices often fail to capture harms affecting edge cases and marginalized groups. This article argues that this failure is structural: the act of “discretization”—converting continuous reality into discrete governance categories—inevitably produces a “residual.” Drawing on German Idealism (Kant, Fichte, Schelling) and continental philosophy (Dilthey, Gadamer, Merleau-Ponty), we reconceptualize residuals not as mere noise but as “surprising facts” that should trigger abductive hypothesis revision. We critique checklist-centered governance as a form of proceduralized auditing that can obscure these residuals. This article makes three key contributions: (i) a structural diagnosis of residual production using systems theory and topology; (ii) a philosophical reconstruction of abductive revision as a hermeneutic necessity; and (iii) an institutional design proposal—specifically, the Residual Ledger and Category Revision Protocols—to operationalize “Open Schema” governance. © 2026 by the author.}, keywords = {abductive reasoning; AI governance; algorithmic auditing; categorization; German Idealism; phenomenology; residuals; schematism}, } @article{skejaasadikudushin2026160, title = {Toward sustainable AI leadership: ethical blind spots, accountability gaps and the CARE governance framework}, author = {Skeja A. and Sadiku-Dushi N.}, year = {2026}, journal = {Leadership and Organization Development Journal}, volume = {47}, number = {2}, pages = {336-354}, doi = {10.1108/LODJ-06-2025-0530}, abstract = {Purpose – This conceptual paper addresses the growing concern of leadership accountability and power asymmetries in AI-integrated organizations. It proposes a structured framework to navigate the ethical, institutional and technical blind spots that often go unnoticed in AI governance. Design/methodology/approach – Drawing on interdisciplinary literature from leadership studies, AI ethics and organizational governance, the paper synthesizes insights to propose a novel theoretical framework (“CARE”) for responsible AI oversight. Findings – The study identifies three dimensions of leadership negligence, technical, ethical and institutional, that create accountability gaps and obscure power dynamics in AI systems. The CARE Framework (Control, Awareness, Responsibility and Evaluation) guides leaders in assessing and addressing these gaps across different sectors. Originality/value – This paper contributes a practical, theory-informed model for ethical AI leadership and provides a foundation for future empirical studies. It bridges fragmented discussions across disciplines and informs both scholars and practitioners aiming to implement trustworthy, power-aware AI systems. © Emerald Publishing Limited}, keywords = {Ethical leadership; Work practices; Work values; Work-related attitudes; Workplace behavior; Workplace support}, } @article{mugodm2026161, title = {Ethical frameworks and predictors of ethical artificial intelligence adoption in Kenya’s health sector}, author = {Mugo D.M.}, year = {2026}, journal = {International Journal of Advanced and Applied Sciences}, volume = {13}, number = {4}, pages = {72-85}, doi = {10.21833/ijaas.2026.04.008}, abstract = {This study used a mixed-methods approach that combined a semi-structured questionnaire with a systematic literature review to examine the factors that influence ethical Artificial Intelligence (AI) adoption in Kenya’s health sector. The aim was to provide evidence to support both policy development and practical implementation. Data were collected from 150 healthcare providers working in healthcare institutions and digital health companies in Kenya and were analyzed using multiple linear regression. The main independent variables were data governance, ethical awareness, regulatory compliance, and organizational accountability. The results showed that all four variables significantly predicted ethical AI adoption. Data governance (β = 0.3157, p < .05) and ethical awareness (β = 0.2415, p < .05) were the strongest predictors, followed by organizational accountability (β = 0.1894, p < .05) and regulatory compliance (β = 0.1128, p < .05). Diagnostic tests confirmed the validity of the regression model (VIF < 3.0, Durbin–Watson = 2.12, p > .05). The findings highlight the importance of strengthening data governance practices, developing human capacity, improving organizational accountability, and ensuring compliance with existing regulatory frameworks to support faster and more ethical AI adoption in the healthcare sector. The paper concludes with policy recommendations that emphasize Afrocentric ethical perspectives inspired by African philosophies such as Ubuntu, capacity building for stakeholders in the healthcare ecosystem on ethical AI, stronger regulatory systems, and efforts to increase public trust in AI-based healthcare solutions. © 2026 The Authors.}, keywords = {Afrocentric ethics; Data governance; Ethical awareness; Organizational accountability; Regulatory compliance}, } @article{lehrersfxiet2026162, title = {What Lessons Should Canada Take from the Design of Public Data Exchanges?}, author = {Lehrer S.F. and Xie T.}, year = {2026}, journal = {Canadian Public Policy}, volume = {52}, number = {S1}, pages = {59-70}, doi = {10.3138/cpp.2025-041}, abstract = {Canadian policy-makers face a fundamental trade-off between protecting the public's privacy and fostering the growth of domestic start-ups that rely on machine learning algorithms for decision-making. Because the performance of such algorithms depends critically on the quality and accessibility of training data, information confined in institutional or sectoral silos represents substantial lost economic and social value. To unlock this value responsibly, governments can play a central role in developing public data exchanges that promote efficiency and innovation while maintaining public trust. This article contrasts the data ex-change frameworks of China, the European Union, and the United States, examining how each jurisdiction manages the tensions between data access and control and between privacy and innovation. We argue that elements of China's approach merit closer consideration as Canada contemplates its own model for commercial data marketplaces that would enable firms to innovate responsibly while providing sufficient safeguards. Canadian Public Policy / Analyse de politiques, 2026. All rights reserved / Tous droits réservés. Authorization to reproduce items for internal or personal use, or the internal or personal use of specific clients, is granted by the journal for libraries and other users registered with the Copyright Clearance Center (CCC): https://www.copyright.com.}, keywords = {AI policy and regulation; algorithmic governance; confidentialité et transparence des données; data exchanges; data privacy and transparency; gouvernance algorithmique; marchés de données d'apprentissage; politique et réglementation en matière d'IA; training data markets; échanges de données; Canada; China; United States; accessibility; algorithm; data quality; European Union; machine learning; trade-off}, } @article{chegen2026163, title = {Is Africa Ready for AI? Digital Information Privacy Awareness and AI Adoption on the Continent}, author = {Chege N.}, year = {2026}, journal = {Social Sciences}, volume = {15}, number = {3}, pages = {155}, doi = {10.3390/socsci15030155}, abstract = {Respect for privacy has been identified as a guiding principle for the development and use of responsible or ethical artificial intelligence (AI), but also as an endangered value in many countries, including those in Africa. Yet, on the African continent, awareness of personal information privacy remains in its early stages, and awareness-raising initiatives are still limited, fragmented, and non-governmental-driven. Given the current global and local enthusiasm surrounding the adoption and development of AI technologies, I examine the key interrelated factors driving the poor digital information privacy awareness and limited awareness-raising in African countries. Key factors include limited digital literacy; the widespread use and reliance on free and freemium services offered by global North digital technology multinationals; the lack of harmonized data protection legislation and regulation across the continent, which facilitates corporate neocolonialism; and the general apathy of many African governments towards privacy awareness-raising, given their own involvement in privacy-violating surveillance. Subsequently, I recommend strategic actions applicable to diverse stakeholders that could contribute towards reinforcing digital information privacy awareness, particularly within the context of the ongoing adoption and anticipated widespread use of AI technologies on the continent. © 2026 by the author.}, keywords = {AI in Africa; artificial intelligence; digital information privacy awareness; ethical AI; privacy; responsible AI}, } @article{dingmdongs2026164, title = {Digentity (Digital-Entity) Human Dyad: A Perspective on Human-AI Collaboration and Decision-Making}, author = {Ding M. and Dong S.}, year = {2026}, journal = {Australasian Marketing Journal}, volume = {34}, number = {2}, pages = {120-125}, doi = {10.1177/14413582251340488}, abstract = {The paper introduces the concept of the Digentity Human Dyad (DHD), a novel paradigm envisioning a future partnership between a human and their digentity—a personalized digital entity that embodies an individual’s values, preferences, and ideals. Unlike digital twins or extended digital selves, digentities are shaped by personal aspirational traits and enriched with the accumulated wisdom of humanity, guiding individuals in their decision-making processes. Enabled by advances in Generative AI, digentities will evolve alongside their human counterparts, provide context-aware and personalized advice, and transform decision-making from a purely human-driven process to a collaborative effort. As digentities align with their human counterpart’s goals, they will influence decisions across various aspects of life, including consumption, personal choices, and societal participation. Beyond individual impact, the DHD has transformative implications influencing how businesses, employers, governments, and civic systems engage with individuals. While the DHD presents significant opportunities, it also introduces challenges, such as risks of bias, data privacy, human over-reliance, and the potential for manipulation. The paper urges organizations and institutions to prepare for these shifts, calling for governance frameworks that ensure responsible AI integration and safeguard human autonomy in an era of human-digentity collaboration. © 2025 Australian and New Zealand Marketing Academy. This article is distributed under the terms of the Creative Commons Attribution-NonCommercial 4.0 License (https://creativecommons.org/licenses/by-nc/4.0/) which permits non-commercial use, reproduction and distribution of the work without further permission provided the original work is attributed as specified on the SAGE and Open Access pages (https://us.sagepub.com/en-us/nam/open-access-at-sage).}, keywords = {Digentity Human Dyad (DHD); Generative AI in decision-making; human-AI collaboration; personalized digital entity}, } @article{effoduhjo2026165, title = {Decolonizing the governance of artificial intelligence in Africa: from normative mimicry to epistemic sovereignty}, author = {Effoduh J.O.}, year = {2026}, journal = {Science and Public Policy}, volume = {53}, number = {2}, pages = {245-257}, doi = {10.1093/scipol/scag005}, abstract = {This article examines the governance of artificial intelligence (AI) in Africa through a decolonial analytical lens that foregrounds the coloniality of power, knowledge, and technology. It argues that emerging national and regional frameworks reproduce structural hierarchies established during colonial rule by translating external governance models into African contexts with limited adaptation to local epistemologies and legal traditions. Through an analysis of the African Union’s Continental AI Strategy and national initiatives in Morocco and Nigeria, the article demonstrates how regulatory mimicry and external dependency constrain Africa’s policy autonomy, while reinforcing epistemic subordination under the guise of digital sovereignty. The study advances a decolonial model of AI governance grounded in epistemic sovereignty, participatory legitimacy, and plural knowledge systems, particularly African philosophical traditions such as Ubuntu and relational personhood. It proposes that contextually attuned governance must integrate indigenous values, community accountability, and regional coordination to ensure that Africa’s digital future is shaped by its own histories, priorities, and normative imaginaries rather than by external conformity. In doing so, the article contributes to a growing body of scholarship on post-colonial technology governance and outlines practical pathways for building inclusive, historically conscious, and self-determined AI governance in Africa. © The Author(s) 2026. Published by Oxford University Press. All rights reserved. For commercial re-use, please contact reprints@oup.com for reprints and translation rights for reprints. All other permissions can be obtained through our RightsLink service via the Permissions link on the article page on our site—for further information please contact journals.permissions@oup.com.}, keywords = {AI governance; artificial intelligence (AI); decolonial approaches; digital sovereignty; epistemic sovereignty; post-colonial theory; technology regulation}, } @article{passadorml2026166, title = {The AI Act's Silent Impact on Corporate Roles}, author = {Passador M.L.}, year = {2026}, journal = {Business Lawyer}, volume = {81}, number = {2}, pages = {393-438}, abstract = {The European Union’s AI Act is poised to reshape corporate governance and compliance well beyond Europe’s borders. By expanding the responsibilities of directors, board secretaries, compliance officers, and in-house counsel, the Act redefines how companies must approach AI oversight, accountability, and risk management. Board secretaries will need to embed AI governance into board procedures, ensuring directors fully understand the risks and opportunities tied to AI deployment. Compliance officers face heightened duties to implement robust risk management frameworks, conduct impact assessments, and manage regulatory reporting obligations. In-house counsel must address the complex allocation of liability, negotiate contractual safeguards, and anticipate conflicts in cross-border compliance regimes. The AI Act’s extraterritorial scope is particularly consequential for US companies: Any AI system that touches the EU market—whether directly developed or simply used in an EU context—triggers regulatory obligations. High-risk AI systems carry stringent requirements around post-market monitoring, transparency, and human oversight. Rather than focusing on granular requirements, this article highlights the AI Act’s structural implications for corporate strategy and legal advisory functions, offering a forward-looking roadmap for mitigating AI-related risks while aligning governance and compliance practices with evolving global standards. For US executives and advisors, the AI Act serves as both a compliance challenge and a governance opportunity: It signals how the European regulation may set the tone for international AI governance and reshape liability, supply chains, and operational models in the transatlantic corporate landscape. In addition to charting regulatory effects, the article makes a distinct legal-methodological contribution: By disaggregating the corporation into its governance roles, it provides the analytical framework that the AI Act itself leaves implicit, rendering its obligations operational within corporate practice. © 2026, American Bar Association. All rights reserved.}, } @article{alhefaitysrsamohamadejamlimritotlarasatiamohamadna2026167, title = {Building trust through Technology: AI, public service perception, and citizen satisfaction in Abu Dhabi policing}, author = {Alhefaity S.R.S.A. and Mohamad E. and Jamli M.R. and Ito T. and Larasati A. and Mohamad N.A.}, year = {2026}, journal = {Multidisciplinary Science Journal}, volume = {8}, number = {4}, pages = {e2026277}, doi = {10.31893/multiscience.2026277}, abstract = {This study examines the impact of artificial intelligence (AI) implementation, perceptions of public services, and trust in government on citizen satisfaction within the Abu Dhabi Police, offering insights into the role of emerging technologies in public administration. Guided by Public Value Theory, Expectation-Confirmation Theory, and IT Assimilation Theory, the research develops a conceptual framework to examine both direct and mediated relationships among these constructs. A purposive sample of 500 police employees from AI-enabled, operational, and administrative units was surveyed using a structured questionnaire, from which 365 valid responses were analyzed to assess the relationships among AI implementation, perception of public services, trust in government, and citizen satisfaction. Data were analyzed using Partial Least Squares Structural Equation Modeling (PLS-SEM) in SmartPLS, enabling a robust assessment of the measurement and structural models. The findings reveal that AI implementation significantly enhances citizens’ satisfaction, particularly when services are transparent, efficient, and aligned with public expectations. Perceptions of public service quality also emerged as a critical determinant of satisfaction, reflecting the importance of accessibility, responsiveness, and fairness in shaping positive citizen experiences. Trust in government was found to play a crucial mediating role, strengthening the link between AI-enabled services, public service perceptions, and satisfaction outcomes. Importantly, the results indicate partial mediation, suggesting that while AI and service quality directly influence satisfaction, their effects are amplified when mediated by trust. These findings highlight the dual importance of technological advancement and institutional credibility in fostering citizen satisfaction. The study contributes theoretically by integrating three complementary frameworks to explain how AI influences service outcomes, and practically by providing evidence-based recommendations for policymakers, AI developers, and law enforcement agencies. Emphasizing transparency, accountability, and ethical AI use can further enhance public trust and maximize satisfaction. This research aligns with the UAE’s Vision 2031 of positioning the nation as a global leader in safety, innovation, and smart governance, while also offering a model for other countries seeking to integrate AI into public service delivery. Copyright (c) 2025 The Authors This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.}, keywords = {artificial intelligence; citizen satisfaction; perception of public services; trust in government}, } @article{jindlksamygnrahimfaselvananthanmmaaropnkrishnanmrperumals2026168, title = {ADAPTING EXISTING PRACTICES FOR ISMS-CERTIFIED ORGANIZATIONS IN SUPPORT OF RESPONSIBLE AI}, author = {Jin D.L.K. and Samy G.N. and Rahim F.A. and Selvananthan M. and Maarop N. and Krishnan M.R. and Perumal S.}, year = {2026}, journal = {ASEAN Engineering Journal}, volume = {16}, number = {1}, pages = {107-124}, doi = {10.11113/aej.V16.23881}, abstract = {Prior to adoption of Artificial Intelligence (AI), organizations may be required to comply with certain industry standards to ensure customer confidence and interoperability of their products, which demands resource allocation and designated responsibilities. For Malaysian public offices certified under the Information Security Management System (ISMS), compliance with a new standard in support of Responsible AI would entail further resources and new reporting structures. Hence, this study proposed the adaptation of current practices for these organizations at the early stages of AI adoption. Ten sources, chosen for authenticity, credibility, representativeness, and meaning, provide the basis for the relevant proposals, including context establishment, risk identification, risk prioritization, and focus area for each control in Annex A of ISO/IEC 27001:2022. The results outlined key actions to support Responsible AI, with future research focusing on validating this framework in ISMS-certified settings. © 2026 Penerbit UTM Press. All rights reserved.}, keywords = {Artificial Intelligence; Framework; ISMS; Responsible AI; Risk Management}, } @article{becchettilsolferinon2026169, title = {Political biases in chatgpt: insights from comparative analysis with human responses}, author = {Becchetti L. and Solferino N.}, year = {2026}, journal = {Economia Politica}, volume = {43}, number = {1}, pages = {285-326}, doi = {10.1007/s40888-025-00384-z}, abstract = {We investigate the political and ideological positioning of ChatGPT, a leading large language model (LLM), by comparing its responses to political economy questions from the European Social Survey (ESS) with those of representative human samples. The questions focus on environmental sustainability, civil rights, income inequality, and government size. We analyze two distinct dimensions of bias: an absolute bias, measured as the deviation of ChatGPT’s answers from the positions of ESS respondents who locate themselves at the center, and a self-perception bias, captured by the difference between ChatGPT’s self-reported left-right placement and the ideological stance which can be inferred from its substantive answers. Our results reveal a significant left-leaning absolute bias in ChatGPT’s responses, particularly on environmental and civil rights issues, which exceeds its own declared center-left self-placement. These findings highlight the importance of transparency regarding AI biases to mitigate unintended ideological influences on users. We conclude by discussing the implications for AI governance, debiasing approaches, and the educational use of language models. © The Author(s) 2025.}, keywords = {ChatGPT; Large language models (LLMs); Political bias}, } @article{lissackm2026170, title = {Rethinking agency in AI ethics: beyond the developer-centric paradigm}, author = {Lissack M.}, year = {2026}, journal = {AI and Society}, volume = {41}, number = {3}, pages = {2157-2166}, doi = {10.1007/s00146-025-02671-6}, abstract = {The rapidly evolving field of artificial intelligence ethics has primarily focused on placing moral responsibility upon technology developers. This article critically examines how the current discourse often misplaces agency by disproportionately assigning ethical responsibility to AI creators while neglecting the roles of users, regulators, and broader societal actors. Drawing parallels to concepts, such as microaggressions and examining key case studies, including the Gebru–Google conflict, this analysis reveals a fundamental issue: prevailing AI ethics frameworks present what are actually constructivist philosophical positions—specifically Rawlsian theories of justice—as if they were realist, objective facts about harm and responsibility. The article demonstrates how claims about “harm” from AI systems often depend on hidden Rawlsian assumptions about unacceptable trade-offs, yet these philosophical choices are rarely made explicit. By interrogating these hidden presuppositions and examining the costs of prioritizing Rawlsian perfectionism over pragmatic innovation, this article argues for a shift from absolutist ethical frameworks toward more realistic, contextual approaches that recognize user agency as primary. Just as hammer manufacturers are not responsible for criminal misuse of their tools, AI developers of general-purpose systems should not bear unlimited responsibility for all possible applications. Only by exposing the constructivist nature of current “ethical” claims and embracing the distributed nature of agency can we develop governance approaches that appropriately assign responsibility across the AI ecosystem. © The Author(s), under exclusive licence to Springer-Verlag London Ltd., part of Springer Nature 2025.}, keywords = {AI ethics; Constructivism; Distributed agency; Pragmatist ethics; Rawlsian justice; Realism; Tool liability; User responsibility; Artificial intelligence; Ethical technology; Product liability; 'current; AI ethic; Constructivism; Distributed agencies; Moral responsibility; Pragmatist ethic; Rawlsian justice; Realism; Tool liability; User responsibility; Economic and social effects}, } @article{xielpengztongx2026171, title = {AI strategy, earnings management, and corporate fraud: Evidence from listed firms in China}, author = {Xie L. and Peng Z. and Tong X.}, year = {2026}, journal = {Economic Modelling}, volume = {156}, pages = {107460}, doi = {10.1016/j.econmod.2025.107460}, abstract = {This study examines how heterogeneous artificial intelligence (AI) strategies affect corporate fraud, addressing a gap in the literature that has largely focused on AI's governance role from a technological perspective while overlooking firms' underlying adoption motivations. Using panel data on Chinese A-share listed firms from 2013 to 2023, we distinguish between symbolic and substantive AI strategies and analyze their differential effects on corporate fraud. The results show that symbolic AI adoption significantly increases fraud risk, particularly among highly financialized firms, non-manufacturing firms, and firms operating under high uncertainty, whereas substantive AI adoption has no direct effect on fraud incidence. Mechanism analysis reveals that symbolic AI increases fraud risk indirectly through accrual-based earnings management, suggesting that opportunistic financial reporting constitutes an important transmission mechanism. In addition, we find that non-standard audit opinions significantly weaken the positive association between symbolic AI adoption and corporate fraud, highlighting the disciplinary role of external audit oversight. Overall, these findings underscore the importance of organisational motivation in shaping the economic consequences of AI adoption and offer policy-relevant implications for fostering more rational and transparent use of emerging technologies. © 2025 Elsevier B.V.}, keywords = {Corporate fraud; Earnings management; Non-standard audit opinion; Substantive AI; Symbolic AI}, } @article{sarkambayevassinghsmukhanovagamralinovabturegeldinovaa2026172, title = {Artificial Intelligence and Business Process Management: A Responsible Framework for Sustainable Transformation}, author = {Sarkambayeva S. and Singh S. and Mukhanova G. and Amralinova B. and Turegeldinova A.}, year = {2026}, journal = {Emerging Science Journal}, volume = {10}, number = {1}, pages = {448-475}, doi = {10.28991/ESJ-2026-010-01-022}, abstract = {This study aims to develop a responsible and sustainable framework for implementing artificial intelligence (AI) in business process management (BPM), with a focus on aligning technological advancement with strategic economic transformation. It addresses the need for ethical, sector-sensitive AI adoption in emerging economies undergoing digital modernization and diversification. The research integrates enterprise information system considerations, privacy-preserving modular architectures, and national regulatory frameworks related to data localization and cybersecurity. A sectoral analysis is conducted to assess global AI adoption maturity and its implications for economic transformation, using Kazakhstan as a contextual reference point. The results reveal that consumer-facing sectors such as retail and financial services exhibit high near-term adoption potential, while healthcare requires gradual infrastructure and talent development. More significantly, mid-term opportunities in manufacturing, logistics, and transportation sectors present Kazakhstan with a comparative advantage. AI adoption in manufacturing is projected to grow by 83% within three to seven years, underscoring the importance of timely investments in automation, smart technologies, and workforce upskilling. This study contributes a context-aware framework for responsible AI-enabled BPM. It offers actionable insights for policymakers and business leaders in emerging economies, advocating for sectoral prioritization, strategic timing, and capacity-building to ensure sustainable digital transformation. © 2026 by the authors.}, keywords = {Artificial Intelligence (AI); Business Process Management (BPM); Business Processes; Data Protection; Data Science; Enterprise Architecture; Ethical AI; Process Improvement; Responsible AI}, } @article{akgnfeakgnm2026173, title = {Governing Generative AI in Healthcare: A Normative Conceptual Framework for Epistemic Authority, Trust, and the Architecture of Responsibility}, author = {Akgün F.E. and Akgün M.}, year = {2026}, journal = {Healthcare (Switzerland)}, volume = {14}, number = {8}, pages = {1098}, doi = {10.3390/healthcare14081098}, abstract = {Background/Objectives: Large language models (LLMs) such as ChatGPT are rapidly being integrated into healthcare for tasks ranging from clinical documentation to diagnostic support. Current ethical discussions focus predominantly on bias, privacy, and accuracy, leaving three critical governance questions unresolved: What kind of knowledge does an LLM output represent in clinical reasoning? When is a clinician’s or patient’s trust in that output justified? Who bears responsibility when an AI-informed decision leads to patient harm? This study proposes the Epistemic Authority–Trust–Responsibility (ETR) Architecture, a normative conceptual framework that addresses these three questions as an integrated governance challenge. Methods: The framework was developed through normative conceptual analysis—a method that constructs governance proposals by synthesising philosophical principles, ethical theories, and empirical evidence. The literature was identified through structured searches of PubMed, PhilPapers, and EUR-Lex (January 2020–March 2026), drawing on the philosophy of medical knowledge, the ethics of trust and testimony, and the moral philosophy of responsibility. Results: The ETR Architecture produces four outputs: (i) a four-tier classification system that distinguishes LLM outputs—from administrative drafts to clinical evidence claims—and matches each tier to appropriate verification requirements; (ii) the concept of the ‘epistemic placebo’, formally defined as a governance measure that creates a documented appearance of compliance while lacking at least one operative element of genuine oversight; (iii) a model specifying four conditions under which trust in healthcare AI is justified; (iv) four testable hypotheses with associated research designs connecting governance design to trust calibration and patient safety. Conclusions: The 2025–2027 regulatory transition period offers a critical window for shaping how healthcare institutions govern AI. We argue that deploying LLMs without explicitly classifying their outputs and building appropriate oversight risks allows governance norms to be set by technology vendors rather than by evidence-informed, patient-centred policy. © 2026 by the authors.}, keywords = {AI governance; automation bias; epistemic authority; generative artificial intelligence; healthcare; human oversight; large language model; patient safety; responsibility gap; warranted trust; placebo; architecture; Article; authority; conceptual framework; ecosystem; epistemology; generative artificial intelligence; health care; health care delivery; human; large language model; patient safety; philosophy; responsibility; trust; witness}, } @article{yinhttangk2026174, title = {Unlocking AI growth in emerging economies: The role of cross-border venture capital}, author = {Yin H.-T. and Tang K.}, year = {2026}, journal = {Pacific Basin Finance Journal}, volume = {98}, pages = {103189}, doi = {10.1016/j.pacfin.2026.103189}, abstract = {AI-investment growth remains constrained by technical complexity and information incompleteness. Leveraging the complementarities of AI venture capital (VC) across broader scopes proves crucial for the AI industry's expansions. We design a state-dependent local projection strategy to investigate the dynamic interplay between cross-border and domestic AI VC in emerging economies, based on a panel dataset spanning 2015–2025. While emerging economies' AI VC is vulnerable to frequent exogenous shocks, it exhibits a symbiotic relationship with cross-border AI VC over the long run. They reinforce and complement each other, but the dynamic impact of cross-border AI VC on domestic AI VC is more pronounced than the reverse in the short and medium run. Improved AI governance, enhanced AI-training dataset accessibility, and relaxed FDI restrictions in fabricated metal industries push the dynamic path of the response difference between non-China emerging economies and China toward zero. Moreover, the response paths triggered by positive and negative changes of cross-border AI VC exhibit asymmetry. Accordingly, we provide policy implications to foster growth in AI VC for emerging economies. © 2026 Published by Elsevier B.V.}, keywords = {Cross-border venture capital; Domestic AI sector; O33; R33; State-dependent local projections}, } @article{hajparkhshinnjkimhs2026175, title = {Differences in perceptions of medical artificial intelligence between medical and non-medical professionals in Korea: a qualitative study}, author = {Ha J. and Park H. and Shinn J. and Kim H.-S.}, year = {2026}, journal = {Journal of the Korean Medical Association}, volume = {69}, number = {3}, pages = {281}, doi = {10.5124/jkma.26.0013}, abstract = {Purpose: Medical artificial intelligence (AI) is rapidly being integrated into clinical practice and healthcare systems, raising concerns regarding safety, accountability, and governance. Despite its increasing importance, empirical comparative studies examining differences in perceptions of medical AI among key expert groups remain limited. This study aimed to compare and analyze perceptions of medical AI among medical and non-medical professionals and to systematically identify commonalities and differences across policy-and governance-relevant domains. Methods: Focus group interviews using open-ended questions were conducted with 30 experts (15 medical and 15 non-medical professionals) who had direct experience with medical AI. Data were analyzed using inductive thematic analysis combined with qualitative comparative analysis. Analytical rigor was strengthened through independent coding and consensus-based discussions. Results: Both groups recognized the potential of medical AI to bring meaningful changes to healthcare systems. However, medical professionals primarily evaluated medical AI in terms of clinical applicability, patient safety, explainability, and accountability. In contrast, non-medical professionals emphasized technological maturity, scalability, data infrastructure, standardization, and system-integration potential. Group-specific patterns also emerged regarding perceived limitations, autonomy, educational priorities, and classification frameworks, particularly in relation to clinical risk management versus system-level design and governance considerations. Conclusion: Differences in perceptions of medical AI are systematically associated with distinct interpretive frames shaped by professional roles and responsibility structures. Effective implementation and policy design for medical AI therefore require an integrated approach that accounts for these structural differences. This study provides empirical evidence and a conceptual foundation for future quantitative and mixed-methods research on medical AI governance. © Korean Medical Association.}, keywords = {Artificial intelligence; Attitude of health personnel; Health policy; Pilot projects; Qualitative research; article; artificial intelligence; clinical practice; consensus; female; health care policy; health care system; health personnel attitude; human; interview; Korea; male; medical society; patient safety; perception; pilot study; professional standard; qualitative research; risk management; standardization; thematic analysis}, } @article{falelakistdimaraaanagnostopouloscn2026176, title = {Systemic Data Bias in Real-World AI Systems: Technical Failures, Legal Gaps, and the Limits of the EU AI Act}, author = {Falelakis T. and Dimara A. and Anagnostopoulos C.-N.}, year = {2026}, journal = {Information (Switzerland)}, volume = {17}, number = {4}, pages = {326}, doi = {10.3390/info17040326}, abstract = {Systemic data bias constitutes a major source of failure in real-world AI systems and represents a regulatory challenge that remains insufficiently addressed by existing legal frameworks, including the EU Artificial Intelligence Act. Although the AI Act introduces a comprehensive risk-based regulatory regime, it does not adequately capture how bias originates, propagates, and manifests across the AI lifecycle. This paper examines systemic data bias through a legal-technical lifecycle analysis that maps recurring bias mechanisms, from data collection and annotation to model training, evaluation, and deployment, to the regulatory control points established under the EU AI Act. Drawing on cross-sectoral examples from employment screening, credit scoring, healthcare risk prediction, biometric identification, and autonomous systems, the analysis demonstrates how technical bias mechanisms translate into systemic governance and accountability challenges. The findings reveal persistent regulatory gaps, including limited auditability of training datasets, the absence of mandatory fairness metrics, insufficient transparency regarding model behavior, and weak mechanisms for post-deployment monitoring and accountability. These results highlight a structural misalignment between lifecycle-based bias dynamics and the Act’s category-driven compliance framework. The paper argues that addressing systemic bias requires a governance approach that integrates technical bias mitigation with legal oversight across the full AI lifecycle rather than relying primarily on post hoc regulatory controls. © 2026 by the authors.}, keywords = {AI governance; algorithmic accountability; EU AI Act; fairness metrics; socio-technical systems; systemic data bias; Artificial intelligence; Biometrics; Data integrity; Laws and legislation; Life cycle; AI governance; AI systems; Algorithmic accountability; Algorithmics; Bias mechanisms; EU AI act; Fairness metric; Real-world; Sociotechnical systems; Systemic data bias; Risk assessment}, } @article{liyyuanmyangyfujxinyduancjwangj2026177, title = {An ethics-informed computable audit framework for monitoring misdiagnosis risk in AI-assisted diagnosis}, author = {Li Y. and Yuan M. and Yang Y. and Fu J. and Xin Y. and Duan C.J. and Wang J.}, year = {2026}, journal = {Scientific reports}, volume = {16}, number = {1}, doi = {10.1038/s41598-026-46652-1}, abstract = {Diagnostic AI can misclassify under distribution shift and subgroup imbalance; governance signals are rarely computable at deploy time. We target deployed diagnostic decision-support systems that perform binary classification and output a continuous risk score (preferably a calibrated probability). The audit layer ingests deploy-time streams including features X, model score S, subgroup tag g, and clinician action a, while outcome labels Y may arrive later via adjudication or follow-up. We define a unit-scaled Misdiagnosis Risk Index (MRI-AI) aggregating shift, fairness, calibration, and human-AI interaction; implement a streaming sentinel with starter bands and stop rules; and log signals and actions in an accountability ledger. A minimal simulation emulates device/site drift and imbalance. Outcomes include deploy-time trigger behavior from label-free indicators and delayed updates of label-dependent metrics (overall/worst-group AUC/FPR, ECE/Brier), as well as trigger rate, top-decile error share, and decision-curve net benefit. Using a controlled, scenario-based synthetic stress-test suite (designed to evaluate the audit/monitoring layer rather than to claim clinical performance of a particular diagnostic model), we report both predictive metrics (overall and worst-group AUC/FPR, ECE, and Brier) and alert-centric endpoints (window-level alert rate, time-to-first-trigger, and persistence). The results show that alert burden remains low under stable conditions and increases in a graded and interpretable manner with shift type and severity, supporting scenario-dependent monitoring and risk-tiered governance actions. Temperature scaling improves calibration while preserving rank-based decision behavior, and subgroup disparities remain explicitly auditable. A computable audit layer-MRI-AI + streaming sentinel + ledger-turns fairness and transparency into actionable controls for diagnostic decision support, enabling auditable monitoring and risk-tiered interventions. © 2026. The Author(s).}, keywords = {AI policy and regulation; Artificial intelligence (AI) diagnostics; Ethical responsibility; Misdiagnosis risk; Patient safety and trust; Artificial Intelligence; Decision Support Systems, Clinical; Diagnostic Errors; Humans; Intelligent Systems; artificial intelligence; clinical decision support system; diagnostic error; ethics; human; prevention and control; smart device}, } @article{stellingalkorenhofpblokv2026178, title = {Making sense of the ‘Human’ in human-centered AI: an Arendtian perspective}, author = {Stellinga L. and Korenhof P. and Blok V.}, year = {2026}, journal = {AI and Society}, volume = {41}, number = {4}, pages = {3083-3094}, doi = {10.1007/s00146-025-02769-x}, abstract = {In recent years, ‘human-centered artificial intelligence’ (HCAI) has emerged as a prominent framing device in the societal debate on the implications of AI. By adopting this phrase, AI ethics discourses make an appeal to the notion of the ‘human,’ while failing to critically reflect on its meaning. Against this background, we pose the question as to what ‘human’ is taken to mean in the context of HCAI. We apply a critical hermeneutic approach to analyze prominent HCAI literature and identify five key characteristics that shape the meaning given to the concept of the ‘human’ in HCAI: universalism, individualism, instrumentalism, psychologism, and exceptionalism. Following this, we introduce Hannah Arendt’s philosophical anthropology to provide an understanding of human existence as essentially political life, and argue that the HCAI discourse would benefit from considering the political dimension of human existence. We conclude the paper by proposing a research program for developing a reconceptualization of HCAI informed by philosophical anthropology. © The Author(s) 2025.}, keywords = {Ethics of AI; Hannah Arendt; Human-centered artificial intelligence; Philosophical anthropology; Zoon politikon; Ethical technology; Ethic of AI; Hannah arendt; Hermeneutic approaches; Human-centered artificial intelligence; Key characteristics; Philosophical anthropology; Political dimensions; Research programs; Zoa politikon; Artificial intelligence}, } @article{fernndezfrincnimbaguirrebm2026179, title = {ETHICS, PERSONAL IMAGE, AND DISINFORMATION IN THE ERA OF DEEPFAKES; [Ética, Imagen Personal y Desinformación en la Era de los Deepfakes]}, author = {Fernández Á.F. and Rincón I.M.B. and Aguirre B.M.}, year = {2026}, journal = {VISUAL Review. International Visual Culture Review / Revista Internacional de Cultura}, volume = {18}, number = {2}, pages = {31-46}, doi = {10.62161/revvisual.v18.5930}, abstract = {The growing significance of digital visual culture raises significant questions regarding image curation, respect for image rights, and the protection of privacy. In this context, the development of deepfake technology exacerbates these issues by enabling unprecedented audiovisual manipulation. This article analyses the ethical, legal, and social implications of the creation and dissemination of deepfakes, emphasizing the challenges posed by identity falsification. It also examines the control and regulatory measures implemented by digital platforms to detect and limit the distribution of such content, highlighting the need for clear ethical guidelines. Furthermore, it addresses the capacity of deepfakes to reinforce biases and discriminatory narratives, undermining trust in visual information and perpetuating harmful stereotypes in the collective imagination. © 2026, VisualCOM Scientific Publications. All rights reserved.}, keywords = {Artificial intelligence; Deepfake; Disinformation; Ethics; Personal image}, } @article{peixlic2026180, title = {How Can Artificial Intelligence Policies Promote the Sustainable Enhancement of Regional Science and Technology Industrial Competitiveness? A Fuzzy-Set Qualitative Comparative Analysis (fsQCA) of Policy Instruments}, author = {Pei X. and Li C.}, year = {2026}, journal = {Sustainability (Switzerland)}, volume = {18}, number = {8}, pages = {4052}, doi = {10.3390/su18084052}, abstract = {The sustainable enhancement of regional science and technology industrial competitiveness is an important objective of artificial intelligence (AI) policy. However, how different AI policy instruments can be combined to achieve this goal remains insufficiently understood. This study aims to address this issue by identifying the configurational pathways through which combinations of AI policy instruments contribute to the sustainable enhancement of regional science and technology industrial competitiveness. Based on a policy instrument framework, we analyze AI policies issued by provincial-level governments in China and apply fuzzy-set qualitative comparative analysis (fsQCA), which is appropriate for examining the combined effects of multiple policy instruments. The results show that no single policy instrument is sufficient to produce high regional science and technology industrial competitiveness. Instead, sustained competitiveness is achieved through multiple equivalent configurations of policy instruments. Three driving pathways are identified—(supply and demand)-environmental resonance, demand-driven (supply-environmental) assurance, and supply–demand complementarity—covering five specific configurations. The variation across configurations indicates that effective AI policy mixes are contingent on regional resource endowments and development conditions. Technology R&D support, talent cultivation and collaboration, and application demonstration and promotion emerge as the most recurrent core conditions across configurations. Accordingly, local governments should develop coordinated AI policy mixes, align differentiated policy pathways with regional conditions, and prioritize technology R&D support, talent cultivation and collaboration, and application demonstration and promotion to sustain long-term regional competitiveness. © 2026 by the authors.}, keywords = {artificial intelligence policy; fsQCA; policy instruments; regional innovation governance; regional science and technology industrial competitiveness; sustainable competitiveness; China; artificial intelligence; comparative study; competitiveness; fuzzy mathematics; governance approach; industrial technology; innovation; local government; qualitative analysis; science and technology}, } @article{qianashawrdabbishlsuhjshenh2026181, title = {Locating Risk: Task Designers and the Challenge of Risk Disclosure in Crowdsourced RAI Content Work}, author = {Qian A. and Shaw R. and Dabbish L. and Suh J. and Shen H.}, year = {2026}, journal = {Proceedings of the ACM on Human-Computer Interaction}, volume = {10}, number = {2}, pages = {CSCW029}, doi = {10.1145/3788065}, abstract = {As AI systems are increasingly tested and deployed in open-ended and high-stakes domains, crowdworkers are often tasked with responsible AI (RAI) content work. These tasks include labeling violent content, moderating disturbing text, or simulating harmful behavior for red teaming exercises to shape AI system behaviors. While prior research efforts have highlighted the risks to worker well-being associated with RAI content work, far less attention has been paid to how these risks are communicated to workers by task designers or individuals who design and post RAI tasks. Existing transparency frameworks and guidelines, such as model cards, datasheets, and crowdworksheets, focus on documenting model information and dataset collection processes, but they overlook an important aspect of disclosing well-being risks to workers. In the absence of standard workflows or clear guidance, the consistent application of content warnings, consent flows, or other forms of well-being risk disclosure remains unclear. This study investigates how task designers approach risk disclosure in crowdsourced RAI tasks. Drawing on interviews with 23 task designers across academic and industry sectors, we examine how well-being risk is recognized, interpreted, and communicated in practice. Our findings highlight the need to support task designers in identifying and communicating risks not only to support crowdworker well-being but also to strengthen the ethical integrity and technical efficacy of AI development pipelines. © 2026 Copyright held by the owner/author(s).}, keywords = {AI red teaming; crowdwork; data annotation; data work; labor; Responsible AI; risk disclosure; well-being; Artificial intelligence; Crowdsourcing; Employment; Ethical aspects; Human engineering; Information systems; Information use; Personnel; AI red teaming; AI systems; Crowdwork; Data annotation; Data work; Red teaming; Responsible AI; Risk disclosure; Well being; Workers'; Occupational risks}, } @article{sunyxsmcowenbweichertjeldardiryhkimdzhuq2026182, title = {Developing a Typology of Roles for STEM-Trained Professionals in AI Policy Engagement}, author = {Sun Y.-X.S. and McOwen B. and Weichert J. and Eldardiry H. and Kim D. and Zhu Q.}, year = {2026}, journal = {Bulletin of Science, Technology and Society}, volume = {46}, number = {1}, pages = {3-16}, doi = {10.1177/02704676251413497}, abstract = {This paper explores the roles of STEM (science, technology, engineering, and mathematics) professionals in AI policymaking, addressing the urgent need for informed governance in emerging technologies. With AI's complex sociotechnical impacts, STEM expertise is crucial for balancing benefits and mitigating risks like bias and privacy concerns. Despite their potential influence, the specific contributions of STEM professionals in AI policy remain underexplored. To address this gap, semi-structured interviews were conducted with 15 STEM professionals who have both educational and professional experience in STEM and policy, and have actively participated in AI policymaking. The data revealed two primary role groups: (1) advisors, gatekeepers, and influencers, who leverage their expertise in STEM and policy to guide stakeholders and actively influence policy decisions, and (2) facilitators and brokers, who facilitate connections and manage the flow of information between stakeholders. This typology highlights the varied contributions of STEM professionals to AI governance and policy development. © The Author(s) 2026}, keywords = {AI ethics; AI policy; policy engagement; role ethics; STEM professionals}, } @article{flewttangw2026183, title = {Between technopolitics and geopolitics: AI and middle-power governance in the United Kingdom and Australia}, author = {Flew T. and Tang W.}, year = {2026}, journal = {Journal of Digital Media and Policy}, volume = {17}, number = {1}, pages = {13-33}, doi = {10.1386/jdmp_00198_1}, abstract = {This article examines a significant paradox within the global AI policy landscape, which is the proliferation of AI policies by nation states vs. the perception that national regulation of global technologies like AI is largely ineffective. While the United States, China and the European Union have been claimed to be ‘Digital Empires’ shaping AI governance, this study focuses on the approaches of two middle-sized nations, the United Kingdom and Australia. Both countries, outside the ‘Digital Empires’, have historically been significant policy agents in the global digital space and are developing distinctive AI policy frameworks. Through a qualitative content analysis of six generative AI (GenAI) policy documents from 2021 to 2024, the article identifies shared policy foundations – commitment to domestic AI sector growth and recognition of AI’s transformative potential – but highlights divergences in leadership aspirations and implementation methodologies. The United Kingdom prioritizes AI research and industrial dominance with aspirational objectives, while Australia focuses on trustworthiness, safety and accountability through practical measures and integration with existing governance instruments. This study contributes to understanding the global AI policy landscape by examining how middle powers navigate AI governance amidst the influence of global tech giants and the major AI powers. © 2026 Intellect Ltd.}, keywords = {AI; digital policy; ethical AI; GenAI; governance; policy analysis; tech regulation; trust}, } @article{jdrzejkad2026184, title = {The AI paradox: How AI dependence may erode accounting expertise; [Paradoks sztucznej inteligencji: jak poleganie na sztucznej inteligencji może prowadzić do erozji wiedzy eksperckiej w rachunkowości]}, author = {Jędrzejka D.}, year = {2026}, journal = {Zeszyty Teoretyczne Rachunkowosci}, volume = {50}, number = {1}, pages = {63-87}, doi = {10.5604/01.3001.0055.6632}, abstract = {Purpose: The paper explores the long-term consequences of artificial intelligence (AI) integration in accounting, focusing on two interrelated challenges: AI models’ lack of transparency and the disruption of expertise development resulting from the automation of foundational tasks. Methodology/approach: The study employs a discursive qualitative approach based on a narrative literature review and conceptual synthesis. It draws on theoretical frameworks of skill acquisition, cognitive apprenticeship, and tacit knowledge to analyse how AI technologies are reshaping professional learning pathways and the development of judgement. Findings: AI models’ opacity undermines interpretability and accountability, creating ethical and regulatory challenges that may hinder compliance with accounting principles. The automation of fundamental accounting tasks limits experiential learning opportunities for early-career professionals, thereby impeding their capacity to exercise critical judgement. When reflectively implemented, AI may also serve as a cognitive scaffold that enhances analytical reasoning and knowledge transfer. However, without such reflection, automation and opacity can interact to form a self-reinforcing cycle that gradually erodes professional expertise. Practical implications: The paper highlights the need for deliberate AI governance, professional education reform, and mentoring systems that balance automation efficiency with human interpretive capacity. Originality/value: By linking algorithmic opacity and skill degradation within a unified theoretical framework, the study conceptualises the AI paradox – the coexistence of efficiency gains and expertise erosion – and offers a foundation for future empirical research. © 2026 Stowarzyszenie Księgowych w Polsce.}, keywords = {accounting; accounting education; artificial intelligence; automation; critical thinking; expertise development}, } @article{anl2026185, title = {Governance of Personal Information Security in the Iteration of Generative AI: From the Perspective of the Technological Evolution of Large Models; [生成式人工智能迭代中的个人信息安全治理：基于大模型技术演进视角]}, author = {An L.}, year = {2026}, journal = {Journal of Library and Information Science in Agriculture}, volume = {38}, number = {4}, pages = {61-70}, doi = {10.13998/j.cnki.issn1002-1248.25-0750}, abstract = {[Purpose/Significance] The rapid advancement of generative artificial intelligence (AI) is driving societal digital transformation, yet it simultaneously poses unprecedented systemic risks to personal information security due to the large-scale, automated, and complex nature of its data processing. Previous research has lacked exploration of governance pathways that consider endogenous technological evolution and specific model iterations. This paper takes the technological evolution of mainstream, large-scale generative AI models, both domestically and internationally as a starting point, and systematically reveals the impact of generative AI on personal information protection principles across the stages of data collection, model operation, and content generation. The focus is on analyzing how technological innovations in China's DeepSeek, including open-source traceability, decision transparency, and flexible deployment, lay the groundwork for risk-graded governance. This study not only broadens the theoretical perspective on AI governance and promotes the formation of a "technology-institution" collaborative governance paradigm, but also offers innovative and actionable insights for building an agile and effective personal information protection system in China amidst the rapid adoption of generative AI. [Method/Process] This study employs a comparative analysis and inductive research approach. First, it systematically compares the core technological differences among mainstream generative AI models, both domestic and international, across three dimensions: model ecosystem, model capabilities, and deployment methods. Through this comparison, it analyzes the challenges generative AI poses to personal information protection at various stages, including data collection, model operation, and content generation. Second, the study systematically examines the differentiated impacts brought about by DeepSeek's technological iterations on personal information security governance. Building on this foundation, the research proposes a comprehensive governance strategy centered on the principles of inclusiveness and prudence, guided by risk grading, and covering all operational stages of generative AI. This strategy emphasizes the critical role of DeepSeek's technical characteristics in supporting the implementation of this framework. [Results/Conclusions] The research indicates that constructing a risk-graded governance system based on the sensitivity of personal information is an effective approach to balancing security and innovation in generative AI. This system emphasizes distinguishing between sensitive and general information during data collection, achieving traceability and purpose control during model operation, and implementing differentiated security safeguards during content generation. With its technical advantages, including open-source traceability, decision transparency, and flexible deployment, DeepSeek provides technical validation and practical possibilities for graded governance. This facilitates the protection of sensitive personal information in high-risk scenarios while simultaneously fostering technological iteration and application innovation in medium- to low-risk contexts. Future research should further incorporate multi-dimensional governance elements such as industry self-regulation, social coordination, and international collaboration. Empirical analysis should also be conducted to test the applicability and effectiveness of the governance framework, thereby gradually developing a well-rounded personal information security governance scheme that adapts to the dynamic evolution of technology. © 2026, Agricultural Information Institute, Chinese Academy of Agricultural Sciences. All rights reserved.}, keywords = {deepseek; generative artificial intelligence; personal information security; risk classification}, } @article{limjeschaeferosavulescuj2026186, title = {Critical Engagement: The Value of Transparency of AI in Healthcare}, author = {Lim J.E. and Schaefer O. and Savulescu J.}, year = {2026}, journal = {Philosophy and Technology}, volume = {39}, number = {1}, pages = {1}, doi = {10.1007/s13347-025-01009-w}, abstract = {Why is transparency important for the use of AI in healthcare? Responses to this question typically claim that transparency is something owed to the patient – because it is a condition for informed consent, legitimacy, accountability to the patient, etc. In this paper, we draw attention to why transparency can be valuable for medical practitioners. We claim that transparent AI models facilitate critical engagement by medical practitioners with AI models that they use. That is, they enable practitioners to assess why AI models make the recommendations they do, think about how those reasons affect their own beliefs and judgments, and make reasoned decisions about whether to maintain or change their own judgments. Via this process, AI models can help medical practitioners to improve their practice in a distinctly valuable way. In turn, this benefits both medical practitioners and their patients. This conclusion has important implications for AI design in healthcare: if AI models are to be used in healthcare, they should be designed in ways which allow medical practitioners to understand how the models arrive at their recommendations, and engage with them critically. © The Author(s) 2025.}, keywords = {AI; AI ethics; Black-box; Critical engagement; Explainability; Healthcare; Interpretability; Opaque; Transparency; XAI}, } @article{mariaehalima2026187, title = {Institutional readiness and resistance to AI-based accounting platforms: evidence from Indonesian local governments}, author = {Maria E. and Halim A.}, year = {2026}, journal = {Transforming Government: People, Process and Policy}, volume = {20}, number = {2}, pages = {248-266}, doi = {10.1108/TG-06-2025-0176}, abstract = {Purpose – This study aims to examine how perceived benefits (PB), challenges (PC) and threats (PT) signal readiness and resistance in adopting Indonesia’s artificial intelligence (AI)-based public sector accounting platforms: the Sistem Informasi Pemerintahan Daerah and Artificial Intelligence for Financial Advisor. Design/methodology/approach – Survey data were collected from 364 Public Sector Financial Officers across 44 local governments. Measurement validity was tested using confirmatory factor analysis, group comparisons applied Mann–Whitney U tests and aggregation indices (rwg, intraclass correlation) assessed institutional-level reliability. Findings – Officers with direct system experience reported stronger recognition of predictive and reporting benefits, alongside heightened awareness of operational challenges and career-related threats. These dual patterns reflect both digital maturity and institutional frictions in AI adoption. Research limitations/implications – This study is constrained by its voluntary, cross-sectional design, which limit causal inference and generalizability. Future studies should adopt longitudinal designs, integrate perceptual with administrative data and examine organizational factors such as leadership, infrastructure and interagency coordination. Theoretically, this study extends New Public Management, the Technology Acceptance Model and Innovation Resistance Theory by operationalizing readiness and resistance through PB, PC and PT. Practical implications – Public managers should embed predictive and reporting gains into budget routines, enforce explainability and workflow standards and align AI-enabled work with competency frameworks and career development pathways. Social implications – The findings highlight that AI reform must align with institutional capacity and professional growth, ensuring accountants in the public sector acquire the skills to work effectively alongside AI systems. These measures strengthen accountability, safeguard discretion and support equitable digital transformation in decentralized governance. Originality/value – By treating PB, PC and PT as perceptual indicators of readiness and resistance, this study demonstrates how frontline experiences inform institutional variation in digital maturity and provide actionable insights for AI governance in the public sector. © 2025 Emerald Publishing Limited}, keywords = {AIFA; Artificial intelligence; Digital governance; Direct system experience; Institutional readiness; Public sector accounting; SIPD}, } @article{abufarharkalzoubikhalsafadiaalsousmmnawasrehaelzubimkalashwalfy2026188, title = {Exploring Physicians' Willingness to Integrate Artificial Intelligence in Clinical Practice: Ethical and Practical Insights From a Jordanian Cross-Sectional Survey}, author = {Abu-Farha R.K. and Alzoubi K.H. and Al Safadi A. and Alsous M.M. and Nawasreh A. and El-zubi M.K. and Al-Ashwal F.Y.}, year = {2026}, journal = {Health Science Reports}, volume = {9}, number = {3}, pages = {e71994}, doi = {10.1002/hsr2.71994}, abstract = {Background and Aims: This study explored the practical perspectives of healthcare professionals in Jordan regarding integrating artificial intelligence (AI) tools into clinical practice and describes their concerns about AI's ethical implications. Methods: The study utilized a cross-sectional, questionnaire-based survey that was conducted with employed physicians in Jordan from April through September 2025. The survey used a validated instrument to assess the participants' AI experience, willingness to adopt AI, practical and ethical concerns associated with AI, and support for the recommended actions. The data were analyzed using descriptive statistics and logistic regression. Results: In this study, 297 physicians participated (median age = 36.0; IQR = 19.0). Around 72% of the participants (n = 214) reported having prior experience with AI, while 50.8% (n = 151) expressed an openness to using AI tools in their clinical practice. Physician concerns about AI included a lack of ability to manage complex cases (n = 216, 72.8%), jeopardizing the physician–patient relationship (n = 204, 68.7%), and diminishing their cognitive ability (n = 210, 70.7%). Other ethical concerns included cultural differences (n = 213, 71.7%), and unclear accountabilities for any errors resulting from using AI (n = 209, 70.4%). Physicians who reported being more willing to adopt AI tools had significantly shorter median ages (adjusted odds ratio [AOR] = 0.971, p = 03) and had prior experience with AI (AOR = 0.262, p < 001) and had daily patient case loads of at least 10 patients (AOR = 1.895, p = 05). Conclusion: While Jordanian physicians recognize AI's benefits, they express significant ethical, practical, and contextual concerns. This study highlights the unique concerns of Jordanian physicians, which differ from those in other countries, and underscores the need for region-specific policies addressing training, cultural adaptation, and regulation to support AI integration in clinical practice. © 2026 The Author(s). Health Science Reports published by Wiley Periodicals LLC.}, keywords = {artificial intelligence; clinical practice; ethics; Jordan; physicians; adult; aged; article; artificial intelligence; clinical practice; cognition; cross-sectional study; doctor patient relationship; female; health care personnel; human; Jordanian; major clinical study; male; physician; questionnaire}, } @article{mcilveenetr2026189, title = {Advancing Small Business Strategy in Turbulent Times: Capabilities, Ethical AI, Customer Retention, Brand Trust, Workplace Culture, and Trade Policy Uncertainty}, author = {McIlveene T.R.}, year = {2026}, journal = {Journal of Small Business Strategy}, volume = {36}, number = {2}, pages = {1-6}, doi = {10.53703/001c.157794}, abstract = {Small and medium-sized enterprises (SMEs) are central to innovation, employment, and community development, yet they face intensifying pressures from technological change, shifting customer expectations, cultural strains, and volatile policy environments. This special issue of the Journal of Small Business Strategy emerged from the inaugural Advances in Business Management Conference held in 2025 and brings together diverse theoretical and empirical perspectives on how small firms build resilience and sustain competitive advantage. The contributing articles examine five interrelated domains: capabilities and resilience, AI and digital transformation, customer retention and brand archetypes, workplace “families” and toxic positivity, and the impacts of tariffs and trade policy uncertainty on SME supply chains. Across these themes, the collection highlights that small-business success depends on the alignment of distinct but complementary capabilities, ethically governed technological systems, relational and symbolic foundations of customer relationships, psychologically healthy cultures, and a stable yet dynamic policy environment. Together, the articles advance small business strategy research and offer actionable insights for owners, educators, and policymakers seeking to support SMEs in an era of rapid technological, economic, social, and regulatory change. © 2026 Small Business Institute. All rights reserved.}, keywords = {artificial intelligence (AI); customer retention and branding; organizational resilience; small business strategy; trade policy uncertainty; workplace culture}, } @article{chenj2026190, title = {Pathways to Green AI: Information Disclosure of Artificial Intelligence Within the ESG Framework of Commercial Entities}, author = {Chen J.}, year = {2026}, journal = {Sustainability (Switzerland)}, volume = {18}, number = {6}, pages = {2922}, doi = {10.3390/su18062922}, abstract = {Strengthening transparency has emerged as a pivotal issue in promoting the responsible development of artificial intelligence (AI). As the prevailing framework for corporate information disclosure, Environmental, Social, and Governance (ESG) reporting shares an inherent synergy with AI governance; both are rooted in the pursuit of sustainable development and the disclosure of specific matters to investors and broader stakeholders. This study analyzes the status of artificial intelligence (AI) information disclosure in the ESG (Environmental, Social, and Governance) reports of listed companies across the United States, Europe, and China, finding that: (1) ESG reports have emerged as a primary channel for business organizations to disclose AI-related information; (2) significant disparities exist in disclosure levels across four key AI-related domains—development, application, manufacturing, and consumption; and (3) disclosure density varies considerably across E, S, and G dimensions, with the Governance (G) pillar exhibiting the most comprehensive information. Based on an empirical analysis of the ESG-AI disclosure framework, this study proposes an optimization scheme for ESG-AI reporting, clearly defining mandatory ESG-AI disclosure obligations for listed companies and employing the “comply or explain” mechanism to balance corporate transparency with operational efficiency while adhering to the “Double Materiality” principle by disclosing model training energy consumption and ecological impacts under Environmental (E) matters, addressing employment, employee training, marketing labeling, and customer privacy under Social (S) matters, and elaborating on corporate AI strategies, risk management protocols, and governance policies under Governance (G) matters. Regarding procedural safeguards, taking China as a case study, centralized disclosure could be implemented through the National Enterprise Credit Information Publicity System, complemented by an assurance system for listed company reports to enhance the accessibility and accuracy of information disclosure. © 2026 by the author.}, keywords = {AI ethics; AI governance; AI risk; corporate information publicity; ESG disclosure; China; Europe; United States; artificial intelligence; business development; corporate strategy; ethics; governance approach; risk assessment; stakeholder; sustainable development}, } @article{giudicipkolesnikovv2026191, title = {SAFE AI metrics: An integrated approach}, author = {Giudici P. and Kolesnikov V.}, year = {2026}, journal = {Machine Learning with Applications}, volume = {23}, pages = {100821}, doi = {10.1016/j.mlwa.2025.100821}, abstract = {We contribute to the field of AI governance with the development of a unified compliance metric that integrates three key dimensions of SAFE Artificial Intelligence: Security, Accuracy, and Explainability. While these aspects are typically assessed in isolation, the proposed approach integrates them into a single and interpretable metric, grounded in a consistent mathematical structure. To develop an integrated framework, the outputs of machine learning models are evaluated under three risk dimensions, that correspond to different input data perturbations: data removal (for accuracy); data poisoning (for security); and feature removal (for explainability). The experimentation of the methodology on both real and simulated datasets shows that the integrated metric improves compliance monitoring and enables a consistent evaluation of AI risks. Copyright © 2025. Published by Elsevier Ltd.}, keywords = {Accuracy; AI governance; AI risk management; Explainability; Responsible AI; SAFE AI; Security; Artificial intelligence; Data accuracy; Information systems; Input output programs; Learning systems; Accuracy; AI governance; AI risk management; Explainability; Integrated approach; Key dimensions; Responsible AI; Risks management; SAFE AI; Security; Risk management}, } @article{mengy2026192, title = {Governing AI virtual anchors in China's live streaming E-commerce ecosystem: Policy challenges and global implications}, author = {Meng Y.}, year = {2026}, journal = {Telecommunications Policy}, volume = {50}, number = {2}, pages = {103109}, doi = {10.1016/j.telpol.2025.103109}, abstract = {The rapid advancement of generative artificial intelligence (AI) has fundamentally reshaped the traditional media value chain, transforming the processes of content production, distribution, and consumption. Among these developments, AI virtual anchors have significantly reduced operational costs and enabled the large-scale creation of content. However, their widespread adoption has also raised complex legal, ethical, and regulatory challenges. This paper investigates the governance of AI virtual anchors from three key dimensions. First, it examines how AI technologies are restructuring the media ecosystem, particularly in the realm of live-streaming e-commerce, by displacing human labour and creating new market dynamics. Second, it examines the associated legal and ethical concerns, including intellectual property disputes, the under-recognized rights of “ghost performers”, risks of misinformation, and consumer protection issues. Third, it evaluates China's evolving governance responses, highlighting both proactive regulatory innovations and ongoing challenges. Starting from platform governance theories, this paper develops a China-specific regulatory narrative and identifies a multi-tiered governance system that involves the government, platforms, and public participation, and reveals the underlying logic that redefines platform roles in China's digital governance architecture. This paper argues that China's evolving governance of AI virtual anchors illustrates a distinct institutional model and aims to situate this experience within global discussions, offering comparative reference points for AI governance, particularly regarding platform responsibility, adaptive regulation, and public participation. © 2025 The Author}, keywords = {AI; China; Live streaming; Media regulation; Value chain; Chains; Consumer protection; Ecosystems; Electronic commerce; Government data processing; Media streaming; Philosophical aspects; Public policy; China; Content consumption; Content distribution; Content production; E-commerce ecosystems; Live streaming; Medium regulation; Production distribution; Public participation; Value chains; Artificial intelligence}, } @article{yemfriginale2026193, title = {Engaging stakeholders, shaping AI ethics: Targeted engagement in corporate AI ethics statements}, author = {Ye M. and Friginal E.}, year = {2026}, journal = {PLOS ONE}, volume = {21}, number = {3 March}, pages = {e0340796}, doi = {10.1371/journal.pone.0340796}, abstract = {Corporate AI ethics statements are being increasingly integrated into sustainability communication to showcase responsible AI performance as part of broader sustainable efforts. This study examines how leading AI companies (developers and adopters) engage with diverse stakeholders through their AI ethics statements. Integrating stakeholder theory with the engagement system, it analyses how Strategic, Responsive, and Operational engagement strategies address the needs of primary and secondary stakeholders. The analysis included 122 English-language statements (265,952 words) from those firms, covering their AI ethics guidelines and policies, press releases, and corporate AI ethics reports. The findings show statistically significant differences in usage frequencies in how primary and secondary stakeholders communicated with targeted engagement strategies. For primary stakeholders, particularly customers and third-party developers, companies frequently used three Operation-targeted strategies to detail data protection measures and AI ethics-related product updates. When addressing corporate management and local regulators, Responsive Acknowledge and Strategic Endorse were more prevalent. In contrast, communication with secondary stakeholders, particularly society, interest groups, and academia, frequently used Strategy-targeted engagement to outline broader ethical AI development plans. These findings inform the development of a stakeholder-oriented engagement model for corporate disclosures on their own trustworthy AI practices. © 2026 Ye, Friginal. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.}, keywords = {Artificial Intelligence; Humans; Stakeholder Participation; academia; article; disclosure; English (language); epidemiology; human; major clinical study; open access publishing; practice guideline; responsible artificial intelligence; artificial intelligence; ethics; stakeholder participation}, } @article{fuxzhangqsunmliujyluy2026194, title = {The impact of artificial intelligence development on enterprise labor investment efficiency}, author = {Fu X. and Zhang Q. and Sun M. and Liu J.Y. and Lu Y.}, year = {2026}, journal = {International Review of Economics and Finance}, volume = {107}, pages = {105157}, doi = {10.1016/j.iref.2026.105157}, abstract = {Artificial intelligence (AI) applications have exerted a profound influence on corporate labor hiring decisions. Taking the establishment of the National Pilot Zones for Next-Generation Artificial Intelligence Innovation and Development as a quasi-natural experiment, this study empirically examines the impact of AI policy on labor investment efficiency using a sample of all A-share non-financial listed companies from 2015 to 2024. The results indicate that the establishment of these pilot zones significantly enhances corporate labor investment efficiency, effectively suppressing both over-investment and under-investment in labor. Regarding the underlying mechanisms, the pilot zones function through three primary channels: deepening AI applications helps improve firms' ability to forecast labor demand and optimize human capital investment decisions; easing financing constraints and reducing financing costs provides the necessary liquidity for firms to flexibly adjust their employment scale and avoid sub-optimal decisions driven by short-term financial pressure; and mitigating various agency costs reduces hiring distortions stemming from managerial self-interest by increasing decision-making transparency and management standardization. Heterogeneity analysis reveals boundary differences in policy effects: the positive impact is more pronounced in firms characterized by lower labor costs, higher talent recruitment intensity, severe information asymmetry, and substantial innovation subsidies, reflecting a logic of compensatory distribution of policy dividends and resource synergies. Furthermore, the study finds an empowerment effect where the pilot zones enhance the labor investment efficiency of firms located within an 80 km radius. These findings provide empirical evidence from the Chinese context for understanding human capital allocation in the AI era. © 2026 The Authors.}, keywords = {Agency costs; AI applications; Artificial intelligence pilot zones; Financing constraints; Labor investment efficiency}, } @article{ayataz2026195, title = {Gender Bias in AI Systems: A Critical Analysis of Regulatory Frameworks and Policy Responses}, author = {Ayata Z.}, year = {2026}, journal = {Bialostockie Studia Prawnicze}, volume = {31}, number = {1}, pages = {135-153}, doi = {10.15290/bsp.2026.31.01.08}, abstract = {The rapid proliferation of artificial intelligence systems has exposed pervasive gender biases that reflect and amplify existing societal inequalities, posing significant threats to gender equality and women’s fundamental rights. This article examines gender bias in AI systems through both theoretical and regulatory lenses, analysing how these biases manifest and can be addressed through comprehensive policy frameworks. The first section provides a systematic literature review exploring how bias becomes embedded in algorithmic systems through biased training data, algorithmic design choices, and broader cultural contexts. The second section examines policy responses, comparing UNESCO’s comprehensive recommendations with the European Union’s Artificial Intelligence Act and referencing the Council of Europe Framework Convention on Artificial Intelligence. This analysis reveals a significant disconnect between aspirational frameworks and practical implementation, demonstrating that existing regulatory approaches inadequately address gender bias in AI and highlighting the urgent need for comprehensive integration of gender equality considerations into AI governance frameworks. © 2026 Zeynep Ayata, published by University of Bialystok.}, keywords = {AI governance; artificial intelligence; EU AI Act; gender bias; gender equality}, } @article{barrenoalcaldesdelsovicenteatriveraherediaa2026196, title = {Algorithmic ethics in corporate contexts: Knowledge mapping for responsible management}, author = {Barreno-Alcalde S. and Delso-Vicente A.-T. and Rivera-Heredia A.}, year = {2026}, journal = {Journal of Management and Organization}, volume = {32}, number = {2}, pages = {632-663}, doi = {10.1017/jmo.2026.10089}, abstract = {The incorporation of algorithmic systems into organizations is reconfiguring decision-making processes and raising new ethical challenges related to transparency, impartiality, and accountability. This study maps the field of algorithmic ethics in organizational contexts through a co-citation–based bibliometric analysis of 1,437 Web of Science publications (search conducted on August 20, 2025). The analysis identifies 12 thematic clusters and reveals a robust intellectual structure, with high modularity (Q = 0.726) and a high weighted mean silhouette value (S = 0.894). The findings highlight the centrality of algorithmic management, responsible artificial intelligence, and explainability, as well as bridging works that connect technical, normative, and management-oriented perspectives. The study advances an integrative conceptual model and a future research agenda that point to the emergence of algorithmic ethics as an institutional logic of organizational governance. For managers, the results underscore the need to embed algorithmic ethics within organizational decision-making and control systems. © The Author(s), 2026. Published by Cambridge University Press in association with Australian and New Zealand Academy of Management.}, keywords = {algorithmic ethics; artificial intelligence; bibliometric analysis; organizational governance; responsible AI}, } @article{margalityravivs2026197, title = {The Politics of Using AI in Policy Implementation: Evidence from a Field Experiment}, author = {Margalit Y. and Raviv S.}, year = {2026}, journal = {British Journal of Political Science}, volume = {56}, pages = {e13}, doi = {10.1017/S0007123425101282}, abstract = {The use of AI by government agencies in guiding important decisions (for example, on policing, welfare, education) has triggered backlash and demands for greater public input in AI regulation. Yet it remains unclear what such input would reflect: general attitudes towards new technologies, personal experience with AI, or learning about its implications. We study this question experimentally by tracking the attitudes of over 1,500 workers whose task assignments were randomly determined by either a human or an AI ‘boss’, with task content and valence also randomized. Across a three-wave panel, we find that personal experience with AI-as-boss affected workers’ job performance but not their attitudes on using AI in public decision making. In contrast, exposure to information about the technology produced significant attitudinal change, even when it conflicted with participants’ prior disposition or direct experience. The results highlight the promise of incorporating public input into AI governance. © The Author(s), 2026. Published by Cambridge University Press.}, keywords = {AI governance; algorithmic decision making; field experiment; public opinion}, } @article{michalismrozgonyik2026198, title = {Governing AI at the international and supranational level: Divergent approaches and implications for media governance}, author = {Michalis M. and Rozgonyi K.}, year = {2026}, journal = {Journal of Digital Media and Policy}, volume = {17}, number = {1}, pages = {81-100}, doi = {10.1386/jdmp_00199_1}, abstract = {Despite the lack of consensus on the definition of artificial intelligence (AI), significant efforts have been made in the past decade to establish governance frameworks addressing AI’s broad economic and societal implications. Adopting a governance perspective and based on extensive documentary and legal policy analysis, this article has two main objectives. First, it analyses emerging international and supranational AI governance frameworks, with particular attention to their guiding principles and mechanisms, focusing on the most significant initial attempts from the OECD, G20/G7, UNESCO and the earliest comprehensive legal instruments from the European Union and the Council of Europe. It finds divergence in priorities and values. Second, the article assesses the implications of these differing AI governance approaches for media governance, focusing on the EU. We argue that, although AI policies do not explicitly target the media, they nonetheless have substantial implications for the media. Specifically, the focus on AI regulation has sidelined key media policy debates, despite escalating challenges to media independence and pluralism. Furthermore, AI regulation is embedded in wider transformations in digital governance, institutional change, new platform regulatory mandates and challenges in aligning media policy objectives within this evolving context. An important observation is that long-established media regulators, traditionally tasked with protecting media freedom and democratic values, are being institutionally displaced within the EU’s emerging digital governance architecture. This transformation signals a deeper epistemological move from sector-specific expertise toward transversal forms of oversight, from media policy to technical standardisation and from public interest to market efficiency, product safety and systemic risk management. The article calls for reimagining media regulators’ mandates to include AI oversight, embedding them in horizontal governance structures and ensuring their active role in international and supranational coordination. Without such integration, media governance risks becoming a legacy function increasingly detached from the technological, institutional and actor configurations shaping today’s public sphere. © 2026 Intellect Ltd.}, keywords = {artificial intelligence (AI) governance; democratic media oversight; EU digital regulation; institutional displacement; international regulatory frameworks; platform regulation}, } @article{saeedkprybutokvr2026199, title = {When utility meets ethics: A stakeholder perspective on agentic information systems delegation}, author = {Saeed K. and Prybutok V.R.}, year = {2026}, journal = {International Journal of Information Management}, volume = {86}, pages = {102976}, doi = {10.1016/j.ijinfomgt.2025.102976}, abstract = {Organizations increasingly use agentic information systems (agentic IS) to achieve performance-driven objectives. However, agentic IS raise ethical concerns, such as fairness, bias, autonomy, transparency, responsibility, and privacy. Over 70 % of the stakeholders, such as users, have expressed concerns about the various business applications of agentic IS. However, limited research has examined how stakeholders evaluate utility versus ethics in delegating tasks to agentic IS. Our results show that stakeholders assess agentic IS for delegation using a combination of rules-based and outcome-based ethical thinking, which values both AI utility and ethical considerations. Moreover, when agentic IS embeds ethical capabilities, stakeholders view both agentic IS and delegating tasks to them as ethically sound. Furthermore, stakeholders consistently prefer agentic IS with human oversight over fully autonomous systems. These insights help organizations optimize efficiency through agentic IS while promoting ethical adoption by respecting stakeholder preferences. This research contributes to the IS literature by integrating ethics into the agentic delegation framework, thereby establishing a foundation for future research in the developing field of agentic IS. © 2025 Elsevier Ltd}, keywords = {Agentic IS delegation; AI ethics; Autonomy; Utility; Ethical technology; Information systems; Agentic IS delegation; AI ethic; Autonomy; Business applications; Ethical concerns; Ethical considerations; Human oversight; Performance-driven; Rule based; Utility; Information use}, } @article{kanabarvkaloyanovak2026200, title = {Securing Generative AI Systems: Threat-Centric Architectures and the Impact of Divergent EU–US Governance Regimes}, author = {Kanabar V. and Kaloyanova K.}, year = {2026}, journal = {Journal of Cybersecurity and Privacy}, volume = {6}, number = {1}, pages = {27}, doi = {10.3390/jcp6010027}, abstract = {Generative AI (GenAI) systems are increasingly deployed across high-impact sectors, introducing security risks that fundamentally differ from those of traditional software. Their probabilistic behavior, emergent failure modes, and expanded attack surface, particularly through retrieval and tool integration, complicate threat modeling and control assurance. This paper presents a threat-centric analysis that maps adversarial techniques to the core architectural layers of generative AI systems, including training pipelines, model behavior, retrieval mechanisms, orchestration, and runtime interaction. Using established taxonomies such as the OWASP LLM Top 10 and MITRE ATLAS alongside empirical research, we show that many GenAI security risks are structural rather than configurable, limiting the effectiveness of perimeter-based and policy-only controls. We additionally analyze the impact of regulatory divergence on GenAI security architecture and find that EU frameworks serve in practice as the highest common technical baseline for transatlantic deployments. © 2026 by the authors.}, keywords = {adversarial machine learning; AI governance; cybersecurity architecture; EU AI Act; generative AI security; large language models; MITRE ATLAS; NIST AI Risk Management Framework; OWASP LLM risks; transatlantic regulation}, } @article{sposatom2026201, title = {Leadership playbooks: a theoretical framework for managing AI’s transformative potential in organizations}, author = {Sposato M.}, year = {2026}, journal = {Leadership and Organization Development Journal}, volume = {47}, number = {2}, pages = {309-321}, doi = {10.1108/LODJ-02-2025-0104}, abstract = {Purpose – This paper develops theoretical frameworks for managing artificial intelligence’s (AI) transformative potential within organizations through leadership playbooks, exploring how organizations can integrate AI effectively, while preserving ethical standards and organizational effectiveness. Design/methodology/approach – A theoretical framework emerges through narrative literature review and analysis of contemporary leadership practices within AI-enabled organizations. Findings – The research demonstrates that successful AI integration demands multifaceted leadership playbooks addressing technical, organizational and ethical dimensions while prioritizing human-centric approaches to technological transformation. Research limitations/implications – This study provides theoretical foundations for understanding how leadership practices must evolve to manage AI integration effectively while preserving organizational effectiveness and ethical standards. Practical implications – Organizations can apply the developed framework to create comprehensive leadership playbooks guiding AI implementation while addressing associated risks and challenges. Originality/value – This paper presents a novel theoretical framework for developing leadership playbooks in AI-enabled organizations, contributing to academic literature and practical organizational management. © Emerald Publishing Limited}, keywords = {Artificial intelligence leadership; Digital leadership; Ethical AI implementation; Leadership development; Organizational transformation}, } @article{metcalft2026202, title = {AI safety and regulatory capture}, author = {Metcalf T.}, year = {2026}, journal = {AI and Society}, volume = {41}, number = {3}, pages = {2451-2466}, doi = {10.1007/s00146-025-02534-0}, abstract = {Researchers, politicians, and the general public support safety regulations on the production and use of AI technology. Yet regulations on new technology are susceptible to the harmful phenomenon of regulatory capture, in which organizations and institutions with economic or political power exert that power to use regulations to unjustly enrich themselves. Only a few authors have tried to raise the alarm about regulatory capture in AI safety and even fewer have described the problem and its implications in detail. Therefore, this paper has three related goals. The first goal is to argue for caution: AI safety is a field with enormous potential for such regulatory capture. Second, this paper explores, in detail, a variety of harms and injustices that captured AI-safety regulations are likely to create. The third goal, in the penultimate section, is to review and critique a few proposals that might mitigate the problem of regulatory capture of AI safety. © The Author(s) 2025.}, keywords = {AI ethics; AI governance; AI safety; Regulatory capture; Laws and legislation; Safety engineering; AI ethic; AI governance; AI safety; AI Technologies; General publics; Power; Public support; Regulatory capture; Safety regulations; Artificial intelligence}, } @article{kaeferfmoraglsantosjc2026203, title = {“Fair” and “Just” Generative Artificial Intelligence for the Base of the Pyramid Population}, author = {Kaefer F. and Mora G.L. and Santos J.C.}, year = {2026}, journal = {Journal of Macromarketing}, volume = {46}, number = {1}, pages = {84-91}, doi = {10.1177/02761467251375981}, abstract = {The rapid growth in Artificial Intelligence (AI), while having positive elements, also raises important ethical concerns, especially concerning vulnerable populations. Recent macromarketing research has indicated that if AI is not used properly, it could lead to greater injustice in the world. In this paper, we focus on Generative AI and present guidelines to achieve fairness and justice for populations at the Base of the Pyramid (BOP), which are known to be particularly vulnerable. These guidelines emphasize the responsibility of organizations to mitigate the detrimental effects of AI, the need to socially innovate AI applications, provide education to vulnerable stakeholders, and to ensure all are represented in the data used to develop AI-based products and services. We use a normative framework labeled the Integrative Justice Model (IJM) to develop these guidelines. We hope that these guidelines can assist in the development of “fair” and “just” GenAI for the BOP. © The Author(s) 2025}, keywords = {AI ethics; artificial intelligence (AI); base of the pyramid (BOP); fairness; generative AI (GenAI); integrative justice model (IJM); justice; vulnerable populations}, } @article{leoijmadunezenwalramkissoonladebayopmugumetnalwoogas2026204, title = {AI Ethical Policy in Africa}, author = {Leo I.J. and Madu N. and Ezenwa L. and Ramkissoon L. and Adebayo P. and Mugume T. and Nalwooga S.}, year = {2026}, journal = {Digital Government: Research and Practice}, volume = {7}, number = {1}, pages = {4}, doi = {10.1145/3776546}, abstract = {Artificial intelligence (AI) is a rapidly growing sector within the African innovation ecosystem. Amidst rapid technological advancements in the development of AI solutions in African countries, it has been noted that AI technologies pose ethical quandaries. As a result, there is a need to identify these AI ethical policies in Africa and explore their benefits and detrimental in the AI ecosystem. Therefore, this study utilized both qualitative and quantitative methodologies using desk research and AI stakeholders' engagement through key information interviews and focus group discussions, case studies, online surveys, and webinar sessions. A total of 300 publications and 165 participants contributed to the study. The result shows that about 12% of the collected responses indicate that the adoption, development, and use of AI ethical policies in African countries are in the initial stage. The AI ethical policies are beneficial; however, the following challenges hinder their adoption and development, which include limited understanding of AI, funding, lack of access to data, inadequate infrastructure, such as internet connectivity, and skills shortage. The study recommends the enhancement of collaborative networks for resource sharing within the AI stakeholders' ecosystem and the development of adequate infrastructure that enhances the adoption of AI ethical policy. © 2026 Copyright held by the owner/author(s).}, keywords = {Additional Key Words and Phrases Artificial Intelligence; Africa; AI Stakeholders; Ethical Policy; Distributed computer systems; Economic and social effects; Ecosystems; Ethical technology; Human engineering; Additional key word and phrase artificial intelligence; Africa; Artificial intelligence stakeholder; Artificial intelligence technologies; Ethical policy; Key words; Key-phrase; Qualitative methodologies; Quantitative methodology; Technological advancement; Artificial intelligence}, } @article{azzuttia2026205, title = {AI governance after MiFID II: beyond (mere) technological neutrality?}, author = {Azzutti A.}, year = {2026}, journal = {ERA Forum}, volume = {27}, number = {1}, pages = {7-31}, doi = {10.1007/s12027-026-00871-1}, abstract = {This article examines the evolving intersections between artificial intelligence (AI) and EU financial regulation, focusing on the Markets in Financial Instruments Directive II (MiFID II). Grounded in the principle of technological neutrality, MiFID II seeks to enhance investor protection, safeguard market integrity, and ensure that innovation develops within competitive and well-regulated markets across the Union. The article argues, however, that while this neutrality renders the framework functionally enabling, it also leaves it normatively silent in the face of the distinctive and evolving risks introduced by financial AI. As AI applications become increasingly heterogeneous—both across the financial functions in which they are deployed and in their underlying lifecycles and value chains—MiFID II’s activity-based logic increasingly struggles to accommodate their diverse and evolving risk profiles. Reflecting the EU’s broader shift toward risk-based AI governance, the article outlines an initial taxonomy of financial AI applications designed to guide the proportionate alignment of regulatory obligations with AI-related risks, thereby supporting the continued adaptability, coherence, and future-proofing of EU financial services law. © The Author(s) 2026.}, keywords = {AI governance; Artificial intelligence; MiFID II; Risk-based regulation; Technological neutrality}, } @article{fangmnguyenvtleminhtlouiejphamlnhewsonc2026206, title = {Leadership networks: Shaping AI innovations through responsible practices in Vietnamese tourism and hospitality firms}, author = {Fang M. and Nguyen V.T. and Le Minh T. and Louie J. and Pham L.N. and Hewson C.}, year = {2026}, journal = {Tourism Management}, volume = {113}, pages = {105317}, doi = {10.1016/j.tourman.2025.105317}, abstract = {Artificial intelligence (AI) holds significant promise for elevating customer experiences and environmental, social and governance performance in the tourism and hospitality (TH) industry. However, its adoption and use raise ethical concerns and risks, necessitating an effective leadership approach to guide responsible AI-driven innovations and mitigate consequences, such as diminished stakeholder trust. This study examines responsible AI diffusion in TH firms in a developing country through Network Leadership and Actor Network Theory (ANT). It also explores how leadership can be applied to influence the AI-driven approach to improve environmental, social, and governance performance. Understanding the fundamental ANT-based network leadership dynamics within Vietnamese firms can be leveraged to explore and exploit innovations under a suitable and responsible AI governance framework. The study presents theoretical and practical implications for leadership and management in the era of leading with AI. © 2025 Elsevier Ltd}, keywords = {Actor network theory; AI for ESG innovations; Diffusion of innovation; Network leadership; Responsible AI; Viet Nam; actor network theory; artificial intelligence; consumption behavior; tourism development; tourism economics}, } @article{kyomenhh2026207, title = {Humane Intelligence in Geropsychiatric Care: Relational Artificial Intelligence, Clinical Wisdom, and the Moral Grid Operational Index}, author = {Kyomen H.H.}, year = {2026}, journal = {American Journal of Geriatric Psychiatry}, volume = {34}, number = {4}, pages = {674-692}, doi = {10.1016/j.jagp.2025.12.012}, abstract = {Artificial intelligence (AI) already influences how older adults are identified for services, supported between provider visits, and referred for care, yet most AI governance currently focuses on algorithms and infrastructure rather than the actual experiences of older adults and their caregivers. Humane Intelligence is a patient-centered, ethically attuned, relational framework for designing, evaluating, and monitoring AI in older adult care. It rests on four pillars: Relational Intelligence, Transparency with Care, Reciprocity and Consent, and Ethical Governance in Strategic Regions, and applies them from point-of-care encounters to system-level decisions. For each pillar, guidelines follow the same sequence: signal a problem, take action, and verify benefit or harm. The framework prioritizes outcomes that matter in geriatric psychiatry, including function, distress, caregiver burden, avoidable utilization, equity, and documented harms or overrides. To protect patients, it draws a firm boundary against fully automated clinical actions and recommends clinical-grade standards for patient-facing programs, including scope-of-action labels, human-in-the-loop safeguards for high-risk situations, postmarket monitoring, and periodic certification. This blueprint aligns with 2025 JAMA Summit on AI priorities, World Health Organization guidance for large multimodal models, United States Food and Drug Administration recommendations for Predetermined Change Control Plans, and Office of the National Coordinator for Health Information Technology decision-support intervention frameworks. The goal is practical: to translate ethics into testable patient-centered routines that clinicians can trust, healthcare systems can implement, and leaders can procure, so that AI augments rather than displaces care. © 2025 American Association for Geriatric Psychiatry.}, keywords = {Aging care; consent; geriatric psychiatry; Humane Intelligence; Moral Grid Operational Index; relational governance; transparency; Aged; Artificial Intelligence; Geriatric Psychiatry; Humans; Patient-Centered Care; aged; aging; algorithm; Article; artificial intelligence; caregiver burden; certification; clinical practice guideline; decision support system; female; Food and Drug Administration; gerontopsychiatry; health care system; human; intelligence; medical informatics; morality; postmarketing surveillance; very elderly; World Health Organization; ethics; person centered care; procedures}, } @article{bonellimi2026208, title = {When AI Disclosure Intensifies: Nonlinear Effects on Governance-Risk Disclosures in Selected U.S. Public Firms}, author = {Bonelli M.I.}, year = {2026}, journal = {Journal of Risk and Financial Management}, volume = {19}, number = {4}, pages = {271}, doi = {10.3390/jrfm19040271}, abstract = {Artificial intelligence (AI) has become increasingly prominent in corporate disclosure, yet its relationship with governance-risk disclosure remains unclear. This study examines whether AI disclosure intensity is nonlinearly associated with governance-risk disclosures among selected U.S. public firms. Drawing on competing governance mechanisms, it argues that rising AI disclosure may initially coincide with heightened control and accountability concerns during periods of organizational and technological transition, but at higher levels may be associated with more stable governance-reporting environments. Using a balanced panel of 53 selected large U.S. public firms observed from 2020 to 2024, the study measures AI disclosure intensity through dictionary-based counts of AI-related terminology in annual Form 10-K filings and captures governance-risk disclosure through references to internal-control weaknesses, restatements, non-reliance statements, and regulatory investigations. Firm and year fixed-effects models with a quadratic specification indicate a robust inverted U-shaped association: governance-risk disclosures rise at low to moderate levels of AI disclosure intensity and decline at higher levels. The findings support a stage-dependent interpretation of AI-related disclosure patterns while underscoring that the evidence is disclosure-based rather than a direct measure of AI governance capability or implementation quality. © 2026 by the author.}, keywords = {artificial intelligence disclosure; corporate governance; financial reporting; governance-risk disclosure; nonlinear effects; U.S. public firms}, } @article{wanggkennedysjohnsonmavellinol2026209, title = {Lessons From One FQHC’s Experience With Artificial Intelligence}, author = {Wang G. and Kennedy S. and Johnson M. and Avellino L.}, year = {2026}, journal = {Journal of Ambulatory Care Management}, volume = {49}, number = {1}, pages = {E31-E38}, doi = {10.1097/JAC.0000000000000541}, abstract = {Objective – The rapid evolution of artificial intelligence (AI) presents opportunities and challenges for health systems, especially safety-net providers like Federally Qualified Health Centers (FQHCs). Safety-net systems may need help with structures and processes for assessing AI applications. To address this need, this article describes Moses-Weitzman Health System’s (MWHS) initial steps toward establishing an AI program that defines intentional and informed AI use. Approach – MWHS established two AI-focused workgroups: one of senior leaders and a cross-departmental group, providing a collaborative space for exploring potential applications, creating guidelines, and discussing concerns. With limited existing templates, MWHS crafted an AI policy emphasizing transparency, privacy, and security, outlining the criteria for implementing AI tools that interact with patient data and ensuring compliance with current regulations. Current AI-related projects focus on automating routine tasks, and research interests include evidence frameworks for making decisions about adopting AI tools and evaluating ambient listening technologies. Findings – Lessons learned in building our AI program are that effective implementation requires tech-savvy leadership, cross-department collaboration, and cautious differentiation between general automation and generative AI. Challenges include the need for agile budgeting, careful vendor vetting, and safe testing environments to assess AI benefits and risks responsibly. Conclusions and Action Steps – MWHS’s AI program underscores a cautious but proactive approach to AI, aiming to balance innovation with operational and ethical considerations, and offers a model for other safety-net systems beginning their AI journeys. © 2025}, keywords = {artificial intelligence; community health centers; organization and administration; primary health care; Artificial Intelligence; Humans; Safety-net Providers; artificial intelligence; human; organization and management; safety net health care}, } @article{cordeirocmadomaitislhuangl2026210, title = {The AI-policy-governance nexus: How regulation and AI shift corporate governance toward stakeholders}, author = {Cordeiro C.M. and Adomaitis L. and Huang L.}, year = {2026}, journal = {Technology in Society}, volume = {84}, pages = {103117}, doi = {10.1016/j.techsoc.2025.103117}, abstract = {This study investigates how artificial intelligence (AI), when deployed under sustainability-oriented policy regimes (e.g., the EU AI Act, CSRD, and the U.S. SEC climate disclosure rule), is catalyzing a shift in corporate governance toward stakeholder accountability. Using a curated corpus of seven open-access regulatory and policy texts, we apply a triangulated approach, corpus linguistics (AntConc) and semantic network analysis (InfraNodus), to map how disclosure, risk, assurance, and stakeholder terms structure the discourse. Robustness checks across three stopword specifications (Spec A/B/C) and phrase-level evidence (N-grams/KWIC) corroborate the centrality of disclosure/report/assurance and the conditional peripherality of transparency/accountability. We propose the AI-Policy-Governance Nexus, a conceptual model explaining how regulatory pressure and AI integration reconfigure governance practices beyond compliance. The findings inform strategy, policy design, and future empirical work on AI-enabled ESG systems. © 2025 The Authors.}, keywords = {Artificial intelligence (AI); Corporate governance; Corpus linguistics; CSRD; Disclosure & assurance; EU AI act; Semantic network analysis; Stakeholder governance; Artificial intelligence; Industrial management; Information systems; Natural language processing systems; Public policy; Regulatory compliance; Semantics; Sustainable development; Artificial intelligence; Corporate governance; Corpus linguistics; CSRD; Disclosure & assurance; Disclosure risk; EU artificial intelligence act; OpenAccess; Semantic network analysis; Stakeholder governance; artificial intelligence; corporate strategy; governance approach; industrial policy; network analysis; regulatory approach; stakeholder; Risk assessment}, } @article{singhjpduamshehua2026211, title = {Diffusion of power and multiplexed governance: evolving networks and clusters for global governance of AI infrastructures}, author = {Singh J.P. and Dua M. and Shehu A.}, year = {2026}, journal = {International Affairs}, volume = {102}, number = {2}, pages = {409-434}, doi = {10.1093/ia/iiag024}, abstract = {Beginning with the United States in 2016, more than 70 countries and international organizations have published strategies and policy recommendations for artificial intelligence (AI) infrastructures. This article locates these policies in the shift from a hierarchical distribution of power to a flatter diffusion of power in which systemic interactions can be top-down, bottom-up or horizontal. A diffusion of power across multiple actors and regions weakens the material and socialization capabilities of hegemonic actors, resulting in global governance outcomes that are described here as ‘multiplexity’. Multiplexity offers a complex and pluralist menu of choices to actors. The computational models employed in this article show complex networks and clusters around multiplex choices that outline patterns of global governance for the evolving AI infrastructures. These networks and clusters cast doubt on many of the extant theories of global governance: those rooted in material power, wherein hegemonic states shape global governance; those where normatively motivated actors shape governance in national contexts; or those where regional patterns (North–South, East–West) are easily discernible. The article locates the origins of multiplexity in a diffusion of power entailing intersecting networks, regions, actors and world-views. There are leaders and great powers in AI, but the rest are not merely followers. In a diffused power scenario, multiple ontologies about the world coexist. The article employs big data mining, specifically latent Dirichlet allocation models from computer science, and process tracing to provide evidence of governance mechanisms for AI. © The Author(s) 2026. Published by Oxford University Press on behalf of The Royal Institute of International Affairs. All rights reserved.}, keywords = {diffusion; evolving; governance; multiplexed; networks}, } @article{olabanjoohonenbergerp2026212, title = {MCDA4AI: A framework for managing n>2 criteria problems in decisions about artificial intelligence}, author = {Olabanjo O. and Honenberger P.}, year = {2026}, journal = {Array}, volume = {29}, pages = {100723}, doi = {10.1016/j.array.2026.100723}, abstract = {Human decisions about the design, deployment, and governance of AI systems are increasingly complex, involving multiple desiderata such as accuracy, fairness, transparency, cost, and environmental impact. This paper proposes a novel and scalable framework for managing this complexity through application of resources from the field of Multi-Criteria Decision Analysis (MCDA). After an introduction to the “n>2 Criteria Problem” in decision-making about AI, an introduction to MCDA methods, and a review of prior applications of MCDA to decision support in decisions involving AI, we lay out our own novel and scalable framework for applying MCDA to such decisions. We illustrate the framework through application to four simulated scenarios involving a medical institution's choice among AI-powered diagnostic tools with varying performance on accuracy, fairness, transparency, and cost. We apply five multi-criteria decision analysis (MCDA) methods — namely, the Weighted Sum Model (WSM), the Analytic Hierarchy Process (AHP), the Technique for Order Preference by Similarity to Ideal Solution (TOPSIS), the VIKOR method (VlseKriterijumska Optimizacija I Kompromisno Resenje), and the Preference Ranking Organization METHod for Enrichment of Evaluations (PROMETHEE) — to rank ten AI models under four distinct weight configurations reflecting realistic deployment contexts. Sensitivity analyses are then conducted to examine how variations in criterion weight assignments affect model rankings across methods and scenarios. In concluding sections we discuss the scalability of these techniques for design, deployment, and policy decisions involving AI, as well as the likely limitations of MCDA methods for decision support involving AI. © 2026 The Author(s)}, keywords = {AI ethics; Artificial intelligence (AI); Decision support; Fairness; Multi-criterial decision analysis (MCDA); Transparency; Analytic hierarchy process; Artificial intelligence; Decision making; Design; Diagnosis; Environmental impact; Sensitivity analysis; Analysis method; Artificial intelligence; Artificial intelligence ethic; Artificial intelligence systems; Decision supports; Fairness; Human decisions; Multi-criteria decision analysis; Multi-criterial decision analysis; Multi-criterial decision analyze; Decision support systems; Transparency}, } @article{keydarrmornshanyyabendo2026213, title = {Bending the Rules: On Large Language Models and Content Moderation}, author = {Keydar R. and Mor N. and Shany Y. and Abend O.}, year = {2026}, journal = {Israel Law Review}, volume = {59}, number = {1}, pages = {96-133}, doi = {10.1017/S0021223726100156}, abstract = {This article examines the transformative impact of large language models (LLMs) on online content moderation, revealing a critical gap between platforms’ rule-based policies and their AI-driven enforcement mechanisms. Using Facebook’s hate speech moderation policies and practices as a case study, we identify a paradox: while content policies are increasingly rule-oriented, AI-driven enforcement seems to operate in a standard-like manner. This disconnect creates transparency, consistency and accountability challenges relating to the delineation of online freedom of expression that are not addressed in the literature, and require attention and mitigation. In this specific context, we introduce the concept of ‘rules by the millions’ to describe how AI systems actually operate through generating vast networks of micro-rules that evade traditional regulatory oversight. This phenomenon disrupts the conventional rules-versus-standards framework used in legal theory, raising urgent questions about the adequacy of current AI governance mechanisms. Indeed, the rapid adoption of LLMs in content moderation has outpaced the human capacity to monitor them, creating a pressing need for adaptive frameworks capable of managing the evolving capacities of AI. © The Author(s), 2026. Published by Cambridge University Press in association with the Faculty of Law, the Hebrew University of Jerusalem.}, keywords = {artificial intelligence; content moderation; large language models; online hate speech; rules v. standards}, } @article{lixliuj2026214, title = {Interpretation of FUTURE-AI, the International Consensus Guideline for Trustworthy and Deployable Artificial Intelligence in Healthcare; [《可信赖且可部署的医疗健康人工智能国际共识指南》FUTURE-AI解读]}, author = {Li X. and Liu J.}, year = {2026}, journal = {Chinese Journal of Epidemiology}, volume = {47}, number = {2}, pages = {207-212}, doi = {10.3760/cma.j.cn112338-20250703-00458}, abstract = {Artificial intelligence (AI) has been widely used in healthcare research, such as disease prevention and diagnosis. However, only a limited number of AI tools have been adopted in health practice due to significant clinical, technical, socio-ethical, and legal challenges. The International Consensus Guideline for Trustworthy and Deployable AI in Healthcare (guideline) was established in 2025 to promote trustworthy and ethical AI in healthcare. The guideline was based on six guiding principles—fairness, universality, traceability, usability, robustness, and explainability (FUTURE-AI), and a set of 30 best clinical practices was defined for operationalization, covering the entire lifecycle of healthcare AI. In this study, we briefly introduced and interpreted the core principles and 30 best practices of the guideline. Our study provides a reference for the design, development, evaluation, and deployment of trustworthy AI tools in healthcare (particularly, epidemiological research) in China. © 2026, Chinese Medical Association}, keywords = {Artificial intelligence in healthcare; FUTURE-AI framework; FUTURE-AI框架; International Consensus Guideline for Trustworthy and Deployable Artificial Intelligence in Healthcare; Interpretation; Artificial Intelligence; China; Consensus; Delivery of Health Care; Humans; artificial intelligence; China; consensus; health care delivery; human}, } @article{mortatimvianamundstockfreitasg2026215, title = {AI in Service Design: A New Framework for Hybrid Human–AI Service Encounters}, author = {Mortati M. and Viana Mundstock Freitas G.}, year = {2026}, journal = {Journal of Service Research}, volume = {29}, number = {1}, pages = {81-96}, doi = {10.1177/10946705251344387}, abstract = {The integration of Artificial Intelligence (AI) in service provision has prompted a revaluation of how we design service encounters, given the emergent role this technology plays in services. This article introduces the concept of the Hybrid Service Encounter to explore the evolving interplay between humans and AI in service contexts. We propose a 2 × 2 framework that categorizes service interactions into four distinct quadrants based on whether the provider and user are human or AI. Drawing on service design literature and current developments in AI, we analyze the implications of these hybrid encounters for service design roles, processes, and outputs. The article further identifies key areas for future research to guide the development of human-centered and ethical AI-powered services. Our contribution extends service design theory and practice by offering a framework to guide the integration of AI in service encounters. © The Author(s) 2025. This article is distributed under the terms of the Creative Commons Attribution-NonCommercial 4.0 License (https://creativecommons.org/licenses/by-nc/4.0/) which permits non-commercial use, reproduction and distribution of the work without further permission provided the original work is attributed as specified on the SAGE and Open Access pages (https://us.sagepub.com/en-us/nam/open-access-at-sage).}, keywords = {artificial intelligence; human–AI interactions; service design; service encounters}, } @article{nezrrrozasmg2026216, title = {Artificial intelligence as an ally for a more transparent and participatory open parliament; [La inteligencia artificial como aliada para un parlamento abierto más transparente y participativo]}, author = {Núñez R.R. and Rozas M.Á.G.}, year = {2026}, journal = {Revista Espanola de la Transparencia}, volume = {23}, pages = {139-182}, doi = {10.51915/RET.421}, abstract = {The rapid digitalization of society and the rise of artificial intelligence (AI) are profoundly transforming the legal-political ecosystem. We study the role of AI in an Open Parliament, describing examples from different countries where these tools increase legislative transparency, citizen participation, and institutional accountability. We then address the ethical and legal challenges posed by the introduction of AI in the parliamentary and democratic arena –such as algorithmic opacity, discriminatory biases, the over-automation of political decisions, as well as data privacy and security issues–and propose principles and guidelines for its responsible use (non-discrimination, respect for fundamental rights, human oversight, algorithmic transparency, etc.). The conclusions underscore the need to balance AI’s opportunities to strengthen the Rule of Law with robust safeguards to prevent the erosion of democratic principles. © (2026), (Association of Professionals and Researchers of Transparency). All rights reserved.}, keywords = {Accountability; AI Ethics; Algorithmic Biases; Artificial Intelligence; Citizen Participation; Open Parliament; PALABRAS CLAVE: Inteligencia Artificial; Parlamento abierto; Participación Ciudadana; Rendición de Cuentas; Sesgos Algorítmicos; Transparencia; Transparency; Ética de la IA}, } @article{ibrahimnmulfordfbatistanavarror2026217, title = {AI models for detecting and generating hate speech: implications for Ethiopian policy}, author = {Ibrahim N. and Mulford F. and Batista-Navarro R.}, year = {2026}, journal = {Science and Public Policy}, volume = {53}, number = {2}, pages = {300-315}, doi = {10.1093/scipol/scag006}, abstract = {Ethiopia’s artificial intelligence (AI) policy represents a significant step forward in national digital governance; our study has, however, identified gaps in areas such as linguistic justice and AI safeguards. We investigate the performance of large language models in detecting hate speech in Ethiopian languages—Amharic, Afaan Oromo, and Tigrigna—and their amenability to produce hate speech. Large language models are less effective at detecting hate speech in non-English contexts and can be easily manipulated to create hate speech, raising serious online safety concerns. Upon careful analysis of the policy, we propose ASPIRE, a series of recommendations for updating the policy to address these concerns: adapting policy to the digital sphere, strengthening linguistic inclusivity, preventing AI misuse, improving infrastructure, resourcing media literacy and training, and emphasising overlaps with hate speech governance. Failure to recognize online harms as integral to AI development leaves a policy vacuum that could undermine long-term development goals. © The Author(s) 2026. Published by Oxford University Press. All rights reserved. For commercial re-use, please contact reprints@oup.com for reprints and translation rights for reprints. All other permissions can be obtained through our RightsLink service via the Permissions link on the article page on our site—for further information please contact journals.permissions@oup.com.}, keywords = {African policy; artificial intelligence; Ethiopia; hate speech; large language models; social media}, } @article{ozornnwakairejsalifudsagoernwobodocwallermksuttons2026218, title = {An integrated approach to gender equality, diversity, and inclusion in the development of artificial intelligence tools in agriculture and food system in Africa}, author = {Ozor N. and Nwakaire J. and Salifu D. and Sagoe R. and Nwobodo C. and Waller M.-K. and Sutton S.}, year = {2026}, journal = {AI and Society}, volume = {41}, number = {4}, pages = {2769-2781}, doi = {10.1007/s00146-025-02722-y}, abstract = {Agriculture in sub-Saharan Africa faces complex challenges, such as low productivity, climate stress, and ongoing social inequalities, particularly affecting women and marginalised groups. Whilst artificial intelligence (AI) holds transformative potential for agriculture and food systems, its development often overlooks these stakeholders, thereby reinforcing existing disparities. This study investigates two AI research initiatives in Nigeria and Uganda that employed a design-by-inclusion approach rooted in gender equality, diversity, and inclusion (GEDI) principles. Through retrospective case studies involving small groups of women and persons with disabilities, we examine how participatory engagement influenced the relevance, usability, and confidence of AI tools amongst users. Drawing on insights from Feminist Human–Computer Interaction (HCI) and Design Justice, our analysis demonstrates that inclusive processes led to significant improvements in participants’ confidence and willingness to engage with AI tools. Based on these findings, we propose a practical framework for developing inclusive AI in agriculture. This work underscores the importance of context-sensitive, participatory design in fostering equitable and effective AI innovations within African agriculture. © The Author(s) 2025.}, keywords = {AI in agriculture; Design-by-inclusion; Diversity and inclusion; Gender equality; Inclusive design; Responsible AI; Agriculture; Human computer interaction; Smart agriculture; Agriculture systems; Artificial intelligence in agriculture; Artificial intelligence tools; Design-by-inclusion; Diversity and inclusion; Food system; Gender equality; Inclusive design; Integrated approach; Responsible artificial intelligence; Artificial intelligence}, } @article{dobsonrstowellmwhittakerr2026219, title = {Maintaining patient trust as artificial intelligence’s role in healthcare grows}, author = {Dobson R. and Stowell M. and Whittaker R.}, year = {2026}, journal = {New Zealand Medical Journal}, volume = {139}, number = {1629}, pages = {94-101}, doi = {10.26635/6965.7122}, abstract = {Patient trust is key to the delivery of healthcare and realisation of artificial intelligence’s (AI) benefits in health. Trust in health institutions and the health professionals working within them directly impacts patient engagement with health services and their health outcomes. Patients want to be able to trust the health system and health services to respect, protect and use their data responsibly to minimise any potential harms. Further, when integrating AI within health services, patients want to be able to trust that this is done with good governance, including the correct approvals and processes, to ensure equitable and safe care. Due to the complexity and fast-changing landscape of AI and the varied levels of AI literacy, trust is arguably even more important. Patients need to be able to trust services to use their health information responsibly and integrate AI in care appropriately regardless of whether they fully under-stand the technology. Through transparency and good AI governance, trust can be built and maintained, but if broken or lost, it will be difficult to repair and will have wider implications. This paper provides recommendations for actions to be taken to build and maintain trust in health institutions within the context of the evolving AI landscape. © 2026, Pasifika Medical Association Group. All rights reserved.}, keywords = {Artificial Intelligence; Delivery of Health Care; Humans; New Zealand; Trust; adult; algorithm; Article; artificial intelligence; clinical practice; coronavirus disease 2019; health care; health care delivery; health practitioner; health service; human; knowledge; medical information; mental health; patient engagement; public health service; semi structured interview; health care delivery; New Zealand; trust}, } @article{arifmzhaolzouxlikchenzaiytianm2026220, title = {Ethics training competencies and leadership enable responsible AI in hospitality and tourism}, author = {Arif M. and Zhao L. and Zou X. and Li K. and Chen Z. and Ai Y. and Tian M.}, year = {2026}, journal = {iScience}, volume = {29}, number = {3}, pages = {115134}, doi = {10.1016/j.isci.2026.115134}, abstract = {Artificial intelligence (AI) is reshaping the service industries, increasing demand for fair, transparent, and ethical systems. This study examined how AI ethics training, employee competencies, organizational governance maturity, and leadership commitment jointly supported responsible AI performance. Survey data from employees in hospitality and tourism organizations were analyzed using structural equation modeling. The findings indicated that ethics training was associated with stronger responsible AI performance through two complementary mechanisms: the development of employee competencies and the institutionalization of governance processes. Leadership commitment further strengthened the relationship between governance maturity and performance, highlighting the role of organizational values in ethical technology use. Together, these results indicated that responsible AI emerged from the alignment of learning practices, institutional structures, and leadership priorities. The study provides practical guidance for service organizations seeking to embed ethical principles into AI use. It contributes to broader discussions about responsible digital transformation and sustainable service innovation. © 2026 The Author(s)}, keywords = {Business; Information science}, } @article{kotbnebejbtaddeom2026221, title = {Ethical, legal, and social challenges of data economy in defence the case of battlefield data}, author = {Kot B. and Nebe J.B. and Taddeo M.}, year = {2026}, journal = {AI and Society}, volume = {41}, number = {2}, pages = {1131-1148}, doi = {10.1007/s00146-025-02610-5}, abstract = {Battlefield data have become a critical asset in contemporary defence. Yet there is a gap in the relevant literature, whilst it addresses various aspects of defence data management—including cybersecurity, interoperability, and decision-making support—it overlooks how these data should be collected, curated, and accessed to enhance the responsible development of AI-enabled defence capabilities. This article addresses this gap first by reviewing existing data policies strategies of NATO and Five Eyes Member States to assess the extent to which they focus on battlefield data, and then by outlining how national defence organisations should manage these data to maximise their strategic value whilst mitigating the attendant ethical, legal, and social risks. We argue that due to their non-rivalrous, artificially excludable nature, battlefield data should be conceptualised as an artificial club good and that national defence organisations have ethical obligations to act as club manager to leverage the potential of these data to develop more robust, reliable and controllable AI defence capabilities. We conclude the analysis proposing two sets of policy recommendations to aid national defence organisations in discharging their responsibilities as club managers for battlefield data. © The Author(s) 2025.}, keywords = {Artificial intelligence, AI Ethics; Battlefield data; Data governance; Data sovereignty; Data-centric defence; Defence; Defence digitalisation; Defence technology, Ethics of Defence; Artificial intelligence; Cybersecurity; Ethical technology; Information management; Artificial intelligence, AI ethic; Battlefield data; Data centric; Data governances; Data sovereignty; Data-centric defense; Defence technology; Defense; Defense digitalization; Defense technology, ethic of defense; Decision making}, } @article{sungkarungsripkiattisins2026222, title = {When Advice Isn’t Trusted: Privacy, Transparency, and Accountability Risks Driving AI Mistrust and Consumer Resistance in Financial Advisory Services}, author = {Sungkarungsri P. and Kiattisin S.}, year = {2026}, journal = {Sustainability (Switzerland)}, volume = {18}, number = {3}, pages = {1354}, doi = {10.3390/su18031354}, abstract = {The application of AI in financial planning services has the potential to enhance universal access to financial services. However, AI still faces common consumer mistrust and resistance, hindering the long-term sustainability of AI-powered financial planning. This research aims to explain why consumers resist AI in financial planning and the mechanisms that lead to this resistance and negative customer behavior. This research developed a conceptual model by integrating the S-O-B-C framework with Innovation Resistance Theory, AI ethical risks, and social influence that influence AI mistrust and intention to resist, which lead to negative outcomes such as negative word-of-mouth and customer disloyalty in the context of digital financial planning services in Thailand. The research collected data from a sample of 420 persons and the data was analyzed using PLS-SEM. The research identified social influence and the risks associated with AI transparency and accountability as primary factors contributing to AI mistrust, whereas privacy risk serves as a more fundamental catalyst for resistance. This resistance contributes to negative word-of-mouth and leads to customer disloyalty. It emphasizes that developing sustainable AI financial advisors must go beyond technically secure design to transparent, accountable, and socially legitimate governance to maintain long-term relationships with customers in the digital financial system. © 2026 by the authors.}, keywords = {AI financial advisory services; AI mistrust; artificial intelligence; customer disloyalty; innovation resistance; innovation resistance theory; negative word of mouth; S-O-B-C; social influence; Stimulus–Organism–Behavior–Consequence; Thailand; accountability; artificial intelligence; financial services; innovation; risk assessment; risk factor; spatiotemporal analysis; sustainability}, } @article{xuyxiepnaeemrmalmugrenihameedzagarwals2026223, title = {Responsible AI and employee service innovation behavior: A sequential mediation model of AI self-efficacy and AI crafting}, author = {Xu Y. and Xie P. and Naeem R.M. and Almugren I. and Hameed Z. and Agarwal S.}, year = {2026}, journal = {Technological Forecasting and Social Change}, volume = {224}, pages = {124470}, doi = {10.1016/j.techfore.2025.124470}, abstract = {While the use of artificial intelligence (AI) has become an effective tool for transforming individuals and organizations, adopting a responsible approach to AI systems is imperative. Drawing on conservation of resources theory and social learning theory, this study examines how responsible AI enhances employees' service innovation behavior via employee AI self-efficacy and employee AI crafting, with a particular focus on the moderating role of leader AI crafting. We tested the proposed relationships using structural equation modeling with data collected from 335 U.S. employees working in various service organizations. The findings demonstrate that the indirect effect of responsible AI on employee service innovation behavior is mediated serially by employee AI self-efficacy and employee AI crafting. Furthermore, leader AI crafting strengthens the positive relationship between responsible AI and employee AI self-efficacy. This study contributes to the AI and management literature by highlighting the importance of responsible AI systems in promoting service innovation behavior among employees. This study addresses both theoretical and practical dimensions, as well as proposing directions for future research. © 2025 Elsevier Inc.}, keywords = {Employee AI crafting; Employee AI self-efficacy; Leader AI crafting; Responsible AI; Service innovation behavior; Artificial intelligence; Artificial intelligence systems; Conservation of resources theories; Effective tool; Employee artificial intelligence crafting; Employee artificial intelligence self-efficacy; Leader artificial intelligence crafting; Responsible artificial intelligence; Self efficacy; Service innovation; Service innovation behavior; artificial intelligence; innovation; numerical model; social theory; theoretical study; Human resource management}, } @article{akbarighatarppappasiopuraosvassilakopouloup2026224, title = {Enacting Responsible AI: A Configurational Analysis of AI Principles in Practice}, author = {Akbarighatar P. and Pappas I.O. and Purao S. and Vassilakopoulou P.}, year = {2026}, journal = {Information Systems Frontiers}, volume = {28}, number = {1}, pages = {61-87}, doi = {10.1007/s10796-025-10618-x}, abstract = {Responsible AI (RAI) entails developing, using, and governing AI in a human-centred way to ensure its trustworthiness and alignment with human values. Organizations attempt to enact RAI through guiding principles that aim to minimize threats such as bias and privacy violations and enhance outcomes such as transparency and fairness. In the empirical study we report in this paper, we share insights about how AI experts view RAI principles, including relationships among them, these principles, and the enactment of RAI principles. Employing a sequential mixed-methods approach, we analyse survey data from 135 AI experts from Europe and America using fuzzy-set Qualitative Comparative Analysis (fsQCA), complemented by follow-up interviews. Our analysis identifies five equifinal combinations of RAI principles associated with high overall RAI enactment. The findings reveal the crucial role of accountability and safety-reliability principles and highlight the contextual variability of other principles. This research contributes a nuanced understanding of RAI enactment, moving beyond individual principles to demonstrate their complex interplay and their contextual dependencies in achieving responsible AI practices. © The Author(s) 2025.}, keywords = {Accountability; Benevolence; Enactment; Explainability; Fairness; FsQCA; Inclusiveness; Privacy; Responsible AI; Safety; Security; Transparency; Artificial intelligence; Accountability; Benevolence; Enactment; Explainability; Fairness; FsQCA; Inclusiveness; Privacy; Responsible AI; Security; Accident prevention; Data privacy; Transparency}, } @article{ganttwboxallaburgwinkeldzarejeddimdjidrovskiifriedrichsshardybhartungthollanddkarwathakienhuisakleinstreuernlinzmarczyloelmarvugliaaqianhvanravenzwaaybreespsarimveishtralautwilmotlzalewskiarouquid2026225, title = {Building trust in the integration of artificial intelligence into chemical risk assessment: findings from the 2024 ECETOC workshop}, author = {Gant T.W. and Boxall A. and Burgwinkel D. and Zare Jeddi M. and Djidrovski I. and Friedrichs S. and Hardy B. and Hartung T. and Holland D. and Karwath A. and Kienhuis A. and Kleinstreuer N. and Lin Z. and Marczylo E.L. and Marvuglia A. and Qian H. and van Ravenzwaay B. and Rees P. and Sarimveis H. and Tralau T. and Wilmot L. and Zalewski A. and Rouquié D.}, year = {2026}, journal = {Archives of Toxicology}, volume = {100}, number = {5}, pages = {2149-2167}, doi = {10.1007/s00204-025-04286-8}, abstract = {Artificial Intelligence (AI) is increasingly influencing chemical risk assessment, enabling faster, more comprehensive, and potentially more ethical assessments. The application of AI in chemical risk assessment refers to both generative and predictive algorithms encompassing machine learning, to analyse complex chemical, biological, and environmental data and provide insights into adverse effect potential for humans and ecosystems. AI systems support the prediction of chemical hazards, exposure levels, and adverse effects by learning from experimental results, mechanistic models, and regulatory datasets, thereby enhancing the efficiency of safety evaluations. In October 2024, ECETOC held an international workshop, with experts from academia, industry, and regulatory bodies, to reflect upon the historical challenges in integrating multidimensional omics technologies into chemical regulation and explore the current capabilities and future potential of AI in toxicology and regulatory science. Discussions emphasised that implementation of Findable, Accessible, Interoperable, and Reusable (FAIR) data principles is not just a best practice but rather a prerequisite for building transparent, reliable, and unbiased AI systems. The reliability of AI in producing scientifically valid and socially responsible outcomes depends fundamentally on the availability of FAIR data. However, ensuring trustworthiness also requires robust governance frameworks that go beyond data and human oversight. Critical enablers of responsible AI in chemical risk assessment are rigorous governance, explainability, fit-for-purpose applications, and human oversight. ECETOC supports the development of flexible and iterative frameworks advancing development, validation, transparency, accountability, and trust in AI applications in chemicals regulation. © Crown 2026.}, keywords = {AI; Explainable AI; FAIR data; Hazard and risk assessment; Toxicology; Trust; Animals; Artificial Intelligence; Hazardous Substances; Humans; Risk Assessment; Toxicology; Trust; Article; artificial intelligence; cell painting; chemical risk assessment; clinical decision making; data integration; data quality; ethics; generative adversarial network; government regulation; hazard assessment; human; image analysis; large language model; legal compliance; multidimensional omics; natural language processing; nonhuman; omics; prediction algorithm; procedures concerning cells; regulatory toxicology; reliability; reproducibility; risk assessment; standardization; toxicology; validation process; animal; dangerous goods; procedures; risk assessment; toxicity; toxicology; trust}, } @article{albaroudiemansourithatamlehmalameera2026226, title = {HitHire: The future of ethical, fair, and sustainable AI recruitment – A governance framework}, author = {Albaroudi E. and Mansouri T. and Hatamleh M. and Alameer A.}, year = {2026}, journal = {Array}, volume = {29}, pages = {100592}, doi = {10.1016/j.array.2025.100592}, abstract = {Artificial Intelligence (AI) is transforming recruitment but remains susceptible to algorithmic bias and environmental inefficiencies. This paper presents HitHire, a pilot fairness- and sustainability-aware AI hiring platform tailored to the Saudi Arabian context and aligned with Vision 2030 goals. HitHire integrates large language models (LLMs), adversarial debiasing, Shapley Additive Explanations (SHAP), and real-time carbon tracking to ensure transparent and equitable candidate ranking. Evaluated on 350 anonymized CVs across four job roles (web development, finance, human resources, and data science) using a 70/20/10 train/test/validation split, HitHire achieves notable improvements in fairness metrics—Statistical Parity Difference (SPD) for gender = 0.0156 and Disparate Impact (DI) for nationality = 1.2387—while maintaining strong predictive performance (F1 = 0.96 compared to a baseline of 0.80). The system achieves over a 40% reduction in operational CO2 emissions, with inference energy consumption of 0.003 kWh per query. In a three-month pilot study involving 23 HR professionals within a large Saudi organization, 87% of participants rated system trust at 4 out of 5 or higher. These findings contribute to national digital ethics strategies such as the Saudi Green Initiative, which emphasizes carbon neutrality and sustainable innovation. © 2025 The Author(s)}, keywords = {Adversarial Debiasing; AI governance; Algorithmic Bias; Ethical recruitment; Explainable AI; Fairness in AI; Human-in-the-Loop systems; Saudi Vision 2030; SHAP explainability; Sustainable AI; Artificial intelligence; Employment; Ethical technology; Personnel; Sustainable development; Adversarial debiasing; Algorithmic bias; Algorithmics; Artificial intelligence governance; De-biasing; Ethical recruitment; Explainable artificial intelligence; Fairness in artificial intelligence; Human-in-the-loop; Human-in-the-loop system; Loop systems; Saudi vision 2030; Shapley; Shapley additive explanation explainability; Sustainable artificial intelligence; Carbon}, } @article{choiybhongpcparkys2026227, title = {Layered Control Architectures for AI Safety: A Cybersecurity-Oriented Systems Framework}, author = {Choi Y.B. and Hong P.C. and Park Y.S.}, year = {2026}, journal = {Systems}, volume = {14}, number = {4}, pages = {447}, doi = {10.3390/systems14040447}, abstract = {As artificial intelligence (AI) systems become increasingly autonomous, scalable, and embedded in critical digital infrastructure, AI safety has emerged as a significant consideration for cybersecurity, system reliability, and institutional trust. Advances in large language models and agentic systems expand the threat surface to include misalignment, large-scale misuse, opaque decision-making, and cross-border risk propagation, while existing debates remain fragmented across technical, ethical, and geopolitical domains. This paper conducts a structured comparative analysis of AI safety perspectives from ten influential thinkers, examining them across five dimensions and reframing their insights through a cybersecurity lens spanning national governance, industry standards, and firm-level design. Building on this synthesis, the study proposes a layered control architecture that organizes technical safeguards, governance mechanisms, and human oversight into a defense-in-depth structure. The framework is conceptual and theory-building, intended to clarify system-level security reasoning and support future empirical refinement across diverse institutional contexts. © 2026 by the authors.}, keywords = {agentic AI; AI governance; AI safety; cybersecurity; human–AI integration; secure AI systems; trustworthy AI; value alignment; Accident prevention; Alignment; Artificial intelligence; Cybersecurity; Decision making; Man machine systems; Network security; Public works; Agentic artificial intelligence; Artificial intelligence governance; Artificial intelligence safety; Artificial intelligence systems; Cyber security; Human–artificial intelligence integration; Intelligence integration; Secure artificial intelligence system; Trustworthy artificial intelligence; Value alignment; Embedded systems}, } @article{damsiomjmaterskasamekmkotlinskamchervyakovatsousamvdgraaarpitamgrciorgerneram2026228, title = {Enabling factors for AI policy-making in the film and AV sectors: The case of small European markets}, author = {Damásio M.J. and Materska-Samek M. and Kotlinska M. and Chervyakova T. and Sousa M.V.D. and Graça A.R. and Pita M. and Grácio R. and Gerner A.M.}, year = {2026}, journal = {Journal of Digital Media and Policy}, volume = {17}, number = {1}, pages = {127-149}, doi = {10.1386/jdmp_00202_1}, abstract = {This article builds on research conducted under the European Union Horizon project CresCine, a three-year initiative aimed at enhancing the international competitiveness and cultural diversity of small European film markets. Departing from the analysis of the current policy landscape for film and audio-visual (AV) in Europe, we show its disjointedness from artificial intelligence (AI) challenges impacting this industry. This regulatory gap needs to be addressed through sector-specific innovation. Our research offers data-driven insights into market trends and challenges on both the offer (producers) and the demand (audience) side. We present first-hand accounts of AI’s disruptive effects on creative professional practices, established film and AV national and EU-level policies and expectations regarding AI-driven policy interventions, particularly in small European markets. We discuss the interplay between existing regulatory frameworks, particularly the Audiovisual Media Services Directive (AVMSD), and the future of AI policy-making in these sectors, as well as how AI-driven regulation and innovation might impact small European countries’ AV industries. This article contributes to AI policy debates in the film and AV sector by discussing tools and strategies to foster inclusive growth and suggesting future avenues for AI policy frameworks that are fair, responsive to market realities and beneficial to small and peripheral European film and AV industries. © 2026 Intellect Ltd.}, keywords = {AI policy; artificial intelligence; Audiovisual Media Services Directive (AVMSD); audiovisual regulation; cultural and creative industries and sectors (CCSI); innovation in film and media production; media audience; small European film markets}, } @article{bhegadev2026229, title = {A viewpoint on trust, ethics, and transparency in AI-driven learning organizations}, author = {Bhegade V.}, year = {2026}, journal = {Development and Learning in Organizations}, volume = {40}, number = {2}, pages = {27-30}, doi = {10.1108/DLO-08-2025-0311}, abstract = {Purpose – This viewpoint examines the critical role of trust, ethics and transparency in AI-driven learning organizations, exploring how algorithmic opacity and ethical ambiguity create barriers to organizational learning effectiveness. Design/methodology/approach – Drawing on social cognition theory and organizational learning frameworks, this conceptual analysis illustrates how trust erosion in AI-mediated environments creates a cyclical pattern of resistance behavior that impedes knowledge sharing and collaborative learning processes. Findings – AI systems’ lack of transparency generates interpretive uncertainty among employees, leading to skepticism, resistance behavior and the formation of negative cognitive schemas that hinder organizational learning cycles. Trust deficits compound these effects through reduced participation in AI-enabled training programs. Research limitations/implications – As a conceptual viewpoint, this paper lacks empirical data, future studies (longitudinal surveys or field experiments manipulating transparency) should test the proposed cycle and cultural boundary conditions. Practical implications – Organizations must establish transparent AI governance frameworks and ethical oversight mechanisms to build employee trust and maximize the effectiveness of AI-driven learning initiatives. Social implications – Ethical AI implementation in learning organizations contributes to broader societal goals of responsible technology adoption and sustainable institutional development and can improve fairness in upskilling access. Originality/value – This work uniquely integrates trust theory with AI governance frameworks to demonstrate how ethical transparency gaps create psychosocial barriers to learning, offering a novel perspective on human-AI interaction in organizational contexts. Unlike many AI-ethics models that prioritize technical compliance, this viewpoint emphasizes psychological pathways from opacity to employee learning behavior. © 2026 Emerald Publishing Limited}, keywords = {AI governance; Employee resistance; Ethical AI; Organizational learning; Psychological contracts; Trust in AI}, } @article{carpenterrehuylerdpatolesrmcwhorterr2026230, title = {Introducing SAFE-AI: A Behavioral Framework for Managing Ethical Dilemmas in AI-Driven Human Resource Practices}, author = {Carpenter R.E. and Huyler D. and Patole S.R. and McWhorter R.}, year = {2026}, journal = {Administrative Sciences}, volume = {16}, number = {2}, pages = {85}, doi = {10.3390/admsci16020085}, abstract = {Organizations increasingly deploy artificial intelligence (AI) in human resource (HR) decision processes to improve efficiency and strategic execution, yet ethical failures persist when principles remain decoupled from everyday workflow enactment. This paper addresses AI-ethics in HR practice by advancing a behavior-first premise: AI-ethics becomes durable organizational practice only when ethical intent is translated into observable routines and cues that employees can interpret as legitimate and consistently enforced. We introduce the Socially Aware Framework for Ethical AI (SAFE-AI), which integrates normative ethical reasoning (consequentialist and deontological logics), social information processing, and socially informed heuristics as a practical translation layer for HR workflows. SAFE-AI specifies three stages of implementation—moving in (initiation), moving through (navigation), and moving out (culmination)—to guide scoping and constraints, feedback-driven interpretation management, and institutionalized accountability. Because enactment depends on the organizational cue environment, leadership behaviors (ethical intent-setting, resourcing, sensegiving transparency, and enforceable accountability) function as necessary conditions for sustained implementation beyond HR-local governance. We conclude with implications for practice and a testable agenda for research focused on implementation fidelity, cue-consistency mechanisms, and boundary conditions across organizational contexts. © 2026 by the authors.}, keywords = {AI-ethics; artificial intelligence; heuristics; human resources; SAFE-AI; social information processing}, } @article{robertsonnoguesa2026231, title = {Automating injustice: a critical analysis of AI-assisted recruitment processes from a disability justice perspective}, author = {Robertson Nogues A.}, year = {2026}, journal = {AI and Society}, volume = {41}, number = {4}, pages = {2761-2768}, doi = {10.1007/s00146-025-02581-7}, abstract = {This paper critically examines the use of AI-assisted tools in HR recruitment, highlighting their role in reinforcing systemic inequalities, particularly through the lens of disability justice. While artificial intelligence (AI) has been positioned as a solution to human bias in hiring, its reliance on algorithmic decision-making often exacerbates discrimination, disproportionately affecting historically marginalized groups. By quantifying human experiences into data, AI recruitment tools obscure bias, reduce workplace diversity, and perpetuate techno-ableist hiring practices. This study explores how AI reproduces ableist narratives under the guise of efficiency, reinforcing power imbalances that privilege normative ideals of productivity and capability. Using an intersectional framework, it critiques the classification mechanisms that transform disability into a penalizing factor within AI-driven hiring processes. Furthermore, it examines how AI recruitment systems embed colonialist data collection practices, leading to discriminatory outcomes that disadvantage disabled candidates. The paper also interrogates the ethics of AI-based surveillance in recruitment, particularly through psychometric testing and facial recognition, which penalize neurodivergent individuals and those with non-standard embodiments. Ultimately, this work argues that AI hiring tools do not eliminate bias but instead reconfigure it into more opaque and unaccountable forms. It calls for critical resistance to techno-solutionism in HR recruitment, advocating for greater transparency, ethical AI development, and the inclusion of diverse user perspectives. By challenging the assumption that AI can rectify hiring bias, this study underscores the need for systemic change that prioritizes equity, rather than automated exclusion, in the workplace. © The Author(s), under exclusive licence to Springer-Verlag London Ltd., part of Springer Nature 2025.}, keywords = {AI; Artificial intelligence; Disability justice; Decision making; Employment; Ethical aspects; Algorithmics; Classification mechanism; Critical analysis; Decisions makings; Disability justice; Hiring practices; Human bias; Power imbalance; Recruitment process; Through the lens; Artificial intelligence}, } @article{liuwchotiavwanglsharmapalbishrindashs2026232, title = {Intelligence by design: Large language model work integration as strategic enablers for supply chain regeneration through digital and cognitive capabilities}, author = {Liu W. and Chotia V. and Wang L. and Sharma P. and Albishri N. and Dash S.}, year = {2026}, journal = {Technological Forecasting and Social Change}, volume = {224}, pages = {124497}, doi = {10.1016/j.techfore.2025.124497}, abstract = {This research investigates how working with large language models improves the regenerative capabilities of supply chains by developing digital process transformation capability and cognitive supply chain capability, under varying levels of organisational digital experimentation culture and artificial intelligence (AI) governance maturity. This study develops and tests a multi-stage capability architecture, introducing new perspectives on about cognitive automation, AI capability settings, and algorithmic affordances. The model is validated through analysis of responses from 281 respondents in knowledge-intensive fields. Empirical research supports the proposed serial mediation, depicting that incorporating large language models in supply chains enhance regenerative capacity through digital process transformation and the reconfiguration of cognitive supply chains. Digital experimentation culture strengthens the relationship between the large language model integration into supply chain and digital process capability, whereas AI governance maturity strengthens the link between such integration and regenerative capability. This research adds to modern theories on algorithmic cognition and capability orchestration in AI-enabled systems, adds depth to digital operations and strategic management research, and demonstrates how large language model integration can create regenerative supply chains. © 2025 Elsevier Inc.}, keywords = {AI governance; Algorithmic affordance; Cognitive automation; Digital transformation; Large language models; Supply chain regeneration; Artificial intelligence; Cognitive systems; Integration; Supply chain management; Supply chains; Affordances; Algorithmic affordance; Algorithmics; Artificial intelligence governance; Cognitive automations; Digital process; Digital transformation; Language model; Large language model; Supply chain regeneration; algorithm; artificial intelligence; automation; technological development; technology adoption; Algorithmic languages}, } @article{bozdaaa2026233, title = {AI as Political Infrastructure: A Diamond Model of Political AI Ethics}, author = {Bozdağ A.A.}, year = {2026}, journal = {Philosophy and Technology}, volume = {39}, number = {1}, pages = {54}, doi = {10.1007/s13347-026-01058-9}, abstract = {Artificial intelligence (AI) should be understood as political infrastructure that governs recognition, allocation, and futurity through learning-based classification and prediction. This raises the question of how AI’s political agency should be grasped when its power operates through infrastructures that prestructure what can be recognized or acted upon. Rather than treating ethics as principles applied to neutral tools, this article defines AI as learning-based governance that organizes visibility, authority, and risk. Accordingly, it frames AI ethics as political and infrastructural, understanding the political as the ethical and institutional ordering of relations among subjects and institutions. Integrating Enrique Dussel’s philosophy of liberation with Benjamin Bratton’s Stack model, it develops the Diamond Model of Political AI Ethics, which reframes AI’s ethical problems as systemic, layered, and historically embedded and grounds the analysis in Dusselian concepts of exteriority and antifetishism. The model translates four closures through which totality endures (matter, relation, being, time) into four justice dimensions (distributive, relational, ontological, temporal) and identifies valves where closure can be reopened through design, regulation, or collective action. The analysis shows how legibility, anticipation, modulation, and subjectivation convert life into computed value across AI-mediated domains. The Diamond Model contributes by locating where political agency can re-enter algorithmic systems, linking justice dimensions to Stack layers, and grounding political AI ethics in exteriority by treating those rendered illegible or uncomputable as epistemic authorities. The article thus reframes AI ethics as a politics of design aimed at keeping computational totalities incomplete. © The Author(s), under exclusive licence to Springer Nature B.V. 2026.}, keywords = {Algorithmic governance; Antifetishism; Artificial intelligence; Epistemic authority; Exteriority; Infrastructure; Liberation philosophy; Political AI ethics}, } @article{rauchfleischajungherra2026234, title = {The politics of artificial intelligence alignment: Public reactions to AI moderation in the case of Google’s Gemini}, author = {Rauchfleisch A. and Jungherr A.}, year = {2026}, journal = {New Media and Society}, doi = {10.1177/14614448261449271}, abstract = {This study tests how a prominent artificial intelligence (AI) product failure influences public attitudes, focusing on Google Gemini’s generation of controversial images. Drawing on the AI alignment literature, we distinguish three moderation goals that differ in how far they depart from data-driven outputs: safety, bias mitigation, and aspirational imaginaries. We use focusing events research to explain how controversies make governance questions salient. In a preregistered experiment with 1756 participants, we tested responses to two image sets: American Founding Fathers (T1) and German soldiers from 1943 (T2). T1 significantly reduced support for bias-related and aspirational moderation and lowered trust in the company, but did not affect safety-based justifications or perceived political alignment. T2 showed the same directional pattern but did not reach significance; pooled results confirmed the main pattern. These findings show that visible product failures can affect public views on AI governance along dimensions most directly implicated by the controversy. © The Author(s) 2026. This article is distributed under the terms of the Creative Commons Attribution 4.0 License (https://creativecommons.org/licenses/by/4.0/) which permits any use, reproduction and distribution of the work without further permission provided the original work is attributed as specified on the SAGE and Open Access pages (https://us.sagepub.com/en-us/nam/open-access-at-sage).}, keywords = {AI controversies; AI governance; alignment; artificial intelligence; chatbot; content moderation; public opinion; survey experiment; trust in technology}, } @article{nagpalgkcottej2026235, title = {Human in the Loop, or Perceived Oversight? The Psychological Inference That Drives AI Credibility}, author = {Nagpal G.K. and Cotte J.}, year = {2026}, journal = {Psychology and Marketing}, doi = {10.1002/mar.70174}, abstract = {Industry practice and AI governance frameworks increasingly treat human oversight of AI systems as a core trust-building mechanism. Yet the specific psychological inference through which human-in-the-loop (HITL) systems build credibility remains unidentified. We address this gap across two experiments in AI-generated investment research. In Study 1, we manipulate the disclosed source of an identical report (human, black-box AI, human-supervised AI, or glass-box AI) and find that source labels exert only small effects on credibility and no effect on investment behavior. Critically, human-supervised AI produces the lowest credibility of any condition, contradicting the assumption that adding a human is always trust-positive. In Study 2, we decompose HITL into five theoretically grounded candidate signals through which perceived source could shape credibility. Perceived oversight adequacy, the inference that “someone competent reviewed this work before I saw it,” emerges as the dominant mediator of the source-to-credibility relationship. The findings reframe the HITL question from whether to add a human to how to signal that the analytical process was sound. © 2026 The Author(s). Psychology & Marketing published by Wiley Periodicals LLC.}, keywords = {accountability; AI-generated content; consumer psychology; human-in-the-loop; oversight adequacy; perceived credibility}, } @article{sasidharbbhimanathamarahatesindoriaddevikmungleantripathidrpatilp2026236, title = {Ethical Implications of AI in Corporate English Communication: An Analytical Study of Managerial Decision Making}, author = {Sasidhar B. and Bhimanatham A. and Rahate S. and Indoria D. and Devi K. and Mungle A.N. and Tripathi D.R. and Patil P.}, year = {2026}, journal = {Journal of Daoist Studies}, volume = {19}, pages = {693-700}, abstract = {Artificial Intelligence (AI) has become a revolutionary figure in the communication of companies, impacting the way companies decide on the topics of discussion and especially the interaction they have with English speakers. AI systems, including chatbots, virtual assistants, automated writers, and language analyzers, improve communication by being efficient, accurate, and responsive within businesses. As AI becomes more widespread, however, there are significant ethical implications to consider in terms of transparency, accountability, data privacy, algorithmic bias, and the risk of diminishing human judgment. The study aims to analyze and discuss both ethical dimensions of the application of AI in corporate English communication and its effects on managerial decision-making. The study investigates the perceptions and responses of managers to ethical issues and the challenges that arise when working for organisational efficiency and trust. The study aims to reveal the need for ethical guidelines, responsible AI governance, and human oversight mechanisms. The results offer insights into the ethical aspects of communication in an AI age, which can be beneficial for managing organizations in a sustainable and responsible manner. © 2026, University of Hawaii Press. All rights reserved.}, keywords = {AI Governance; Algorithmic Bias; Artificial Intelligence; Business English; Corporate Communication; Data Privacy; Ethical Implications; Managerial Decision Making; Organizational Ethics}, } @article{gamitomc2026237, title = {Harmonising Consensus: The (Geo)political Economy of Standardisation in the AI Act}, author = {Gamito M.C.}, year = {2026}, journal = {Journal of Intellectual Property, Information Technology and E-Commerce Law}, volume = {17}, number = {1}, pages = {75-93}, abstract = {This paper examines the broader implications of current global (geo)political dynamics on AI standardisation and how, in turn, standardisation shapes those dynamics. Focusing on the European Union’s AI Act, it argues that the Regulation’s reliance on delegated technical standards embeds structural and context-dependent vulnerabilities that constrain its ability to project normative influence. Through insights from the political economy of standardisation, the paper explores the interactions between EU and non-EU aspiration and expectations from standards. It contrasts the EU, the US and China approaches and highlights the operational and normative frictions that arise in standard-setting. The findings show an important gap between the EU’s regulatory ambitions and the realities of the global standardisation ecosystem, which ultimately raises questions about the long-term coherence and influence of the AI Act in shaping the global AI value chain. © 2026 Marta Cantero Gamito}, keywords = {AI Act; AI Governance; AI Standardisation; Co-Regulation; Digital Sovereignty; EU Law; Geopolitics of Technology; Harmonised Standards; Strategic Autonomy}, } @article{butttaiqbalmarshadn2026238, title = {From Policy to Pipeline: A Governance Framework for AI Development and Operations Pipelines}, author = {Butt T.A. and Iqbal M. and Arshad N.}, year = {2026}, journal = {IEEE Access}, volume = {14}, pages = {1373-1397}, doi = {10.1109/ACCESS.2025.3647479}, abstract = {Artificial intelligence systems increasingly operate in high-risk domains where regulatory frameworks such as the EU AI Act, NIST AI RMF, and ISO/IEC 42001 impose explicit evidence and accountability requirements. However, existing engineering practice remains largely manual, retrospective, and decoupled from operational pipelines, resulting in inconsistent provenance, limited reproducibility, and inadequate clause-level traceability. This paper introduces Governance as Evidence for AI Pipelines (GEAP), a pipeline-native governance framework that expresses regulatory and organizational policies as machine-interpretable Governance as Code rules. GEAP integrates governance directly into a unified SDLC–MLOps execution spine by enforcing promotion decisions at five gates—Data, Training, Validation, Release, and Operations—each of which emits signed, content-addressed artifacts into a tamper-evident Evidence Backbone. These artifacts are assembled into a per-run Conformity Bundle, from which the proposed Clause-to-Artifact Traceability mechanism deterministically renders clause coverage across multiple regulatory regimes without manual crosswalks or duplicated documentation. The framework further introduces quantitative governance metrics that measure adequacy, completeness, stability, and evidence hygiene. A detailed synthetic case study of an intensive-care sepsis early-warning system demonstrates GEAP’s ability to standardize promotion control, detect policy violations, and produce replayable, audit-ready compliance manifests in a high-risk clinical context. The results show that governance can operate as a deterministic, reproducible, and verifiable pipeline property rather than an external documentation exercise, enabling more disciplined, transparent, and accountable AI deployment practices. © 2013 IEEE.}, keywords = {AI governance; compliance automation; EU AI Act; evidence management; GEAP; ISO/IEC 42001; MLOps; Artificial intelligence; Compliance control; ISO Standards; Regulatory compliance; AI governance; Artificial intelligence systems; Compliance automation; Development and operations; EU AI act; Evidence management; Governance as evidence for AI pipeline; ISO/IEC; ISO/IEC 42001; MLOp; Pipelines}, } @article{schelblebgflathmanncalyhlyonsjbmcneesen2026239, title = {Human or AI Advice? Examining Trust, Influence, and Responsibility in Ethically Charged Human-AI Team Decision-Making}, author = {Schelble B.G. and Flathmann C. and Aly H. and Lyons J.B. and McNeese N.}, year = {2026}, journal = {Journal of Cognitive Engineering and Decision Making}, doi = {10.1177/15553434261434613}, abstract = {AI systems are rapidly being placed in positions where they can directly influence decisions that require a strong consideration of ethics. This article reports on an experiment in which humans worked with a teammate acting as an expert advisor with expert-level knowledge, who informed and influenced their ethical decision-making. The identity of this expert advisor was manipulated to be either human or AI, and the influence exerted by the expert advisor was manipulated to be either low or high. Further, participants completed four scenarios, each exploring a different ethically charged decision. Results indicated that working with an AI teammate expert advisor led participants to perceive significantly lower levels of stress and responsibility when making ethical decisions, but AI expert advisors were also perceived as performing worse than human expert advisors. Additionally, when expert advisors exerted high influence levels, participants felt significantly less stress during the decision-making process. Finally, the scenario and ethical decisions made by participants had pervasive effects on trust, trustworthiness, perceived performance, and perceived power for both human and AI expert advisors. Future research efforts must ensure that the use of AI expert advisors in human-AI teams does not reduce the responsibility humans bear when making various ethical decisions. © 2026, Human Factors and Ergonomics Society}, keywords = {AI ethics; decision-making; human-AI Teaming; responsibility; responsible AI; trust; Artificial intelligence; Ergonomics; Ethical technology; AI ethic; AI systems; Decisions makings; Ethical decision making; Human expert; Human-AI teaming; Responsibility; Responsible AI; Team decision; Trust; adult; article; decision making; ergonomics; ethical decision making; human; responsibility; responsible artificial intelligence; trustworthiness; Decision making}, } @article{kumardsahuck2026240, title = {Artificial intelligence governance and social policy divergence: a comparative political economy perspective on global AI regulation}, author = {Kumar D. and Sahu C.K.}, year = {2026}, journal = {International Journal of Sociology and Social Policy}, pages = {1-25}, doi = {10.1108/IJSSP-02-2026-0129}, abstract = {Purpose – To explain why artificial intelligence (AI) governance produces systematically divergent social policy outcomes across regions despite widespread convergence around ethical standards for AI. The study examines how political–economic institutions shape the allocation of social risks, regulatory authority, labour integration and ethics institutionalisation in AI governance, thereby driving these differences. Design/methodology/approach – This study uses a comparative qualitative policy analysis based on 24 key AI policy documents published between 2018 and 2025 across the European Union, United States, China, and Indo-Pacific economies. Guided by theory, the documents are systematically coded across four institutional dimensions and converted into simple indices to compare governance approaches across the regions. Findings – The findings show clear and systematic differences in how regions govern AI. Five distinct governance models emerge: rights-based (EU), market-driven (US), state-centric (China), hybrid (Australia–Japan–Singapore) and developmental (India). Although many regions use similar ethical language, substantial differences persist in risk allocation, regulatory enforcement, welfare integration and social protection. These differences reflect the historically embedded political–economic institutions shaping each regime. Originality/value – The paper reframes AI governance as a form of social policy shaped by political and economic institutions. It develops a new, evidence-based typology of AI governance models and shows that differences across countries are driven by institutional structures and not by ethical principles alone. © Deepak Kumar and Chinmaya Kumar Sahu}, keywords = {AI regulation; AI social policy; AI social risks; Artificial intelligence governance; Comparative political economy; Technology governance}, } @article{iongmt2026241, title = {Ethical and Legal Challenges of Partially and Fully Autonomous AI in Healthcare: Reinterpreting Liability and Preserving Trust}, author = {Iong M.T.}, year = {2026}, journal = {Bioethics}, doi = {10.1111/bioe.70137}, abstract = {Artificial intelligence (AI) is increasingly embedded in healthcare in a variety of ways, ranging from semi-autonomous decision support systems to the various visions of completely autonomous clinical systems. This article explores the ethical and legal issues that accompany this shift and argues that traditional ways of understanding medical responsibility and the physician-patient relationships have been strained in light of the increasing autonomy of AI. Ethically, AI raises very acute concerns around transparency and informed consent in the case of “black-box” systems, the persistence and amplification of bias, privacy risks arising from data-intensive uses in system development, as well as the harmful effects of automation bias and professional de-skilling. The article also brings to foreground an underexplored concern—the weakening of relational trust in contexts where clinical judgment and interactions are mediated by an AI, particularly where human oversight of the situation is reduced. Legally, this article outlines challenges of liability across the autonomy spectrum and criticizes proposals that would see AI as an independent liable agent. It promotes a human-accountability model that includes clinical responsibility for assistive AI, coupled with enterprise measures of liability and post-market oversight for ever-more autonomy, backed by mutually appropriate insurance and governance measures. The article ends with some regulatory and institutional recommendations in order to facilitate safe, ethical, and equitable integration of AI in healthcare. © 2026 The Author(s). Bioethics published by John Wiley & Sons Ltd.}, keywords = {artificial intelligence; autonomous medical systems; healthcare ethics; liability and accountability; relational trust; article; artificial intelligence; automation bias; bioethics; decision making; decision support system; doctor patient relationship; health care; human; informed consent; medical ethics; privacy; trust}, } @article{stahlbcalirhitchamlliebherrmmurrayrvallejosepyankoukayaa2026242, title = {Big tech, the state, or you? Who is responsible for generative AI addiction?}, author = {Stahl B.C. and Ali R. and Hitcham L. and Liebherr M. and Murray R. and Vallejos E.P. and Yankoukaya A.}, year = {2026}, journal = {Behaviour and Information Technology}, doi = {10.1080/0144929X.2026.2663509}, abstract = {The rise of generative AI introduces significant social and ethical challenges, notably the potential for addiction, mirroring problematic social media and gaming use. Preliminary evidence suggests harmful addictive properties that may lead to individual, social, and societal harms. This article addresses the research questions: Who or what can be held responsible for generative AI addiction, and what consequences can arise from the resulting responsibilities? We first conceptualise generative AI addiction and review initial empirical evidence for problematic use, then employ the conceptual lens of responsibility, reviewing precedents from other addiction areas (substance and behavioural) to inform the governance of generative AI. The paper proposes a model for a responsible AI ecosystem where accountability is distributed across four key stakeholder groups: governmental bodies, tech companies, researchers, and civil society. We argue that tech companies, who stand to gain most, hold a key role and must engage proactively to mitigate risks, cautioning that reluctance to accept these responsibilities may incur severe reputational and financial costs, similar to those faced by the tobacco and pharmaceutical industries. This work contributes to the technology addiction and AI governance discourses by proposing collaborative accountability structures for emerging technologies. © 2026 Informa UK Limited, trading as Taylor & Francis Group.}, keywords = {addiction; Generative AI; responsibility; Artificial intelligence; Costs; Ethical technology; Addiction; Civil society; Financial costs; Generative AI; Property; Research questions; Responsibility; Social gaming; Social media; Stakeholder groups; Engineering research}, } @article{bernasconilgrossmannr2026243, title = {Navigating the future of clinical trial management – insights on the transformative role of AI}, author = {Bernasconi L. and Grossmann R.}, year = {2026}, journal = {Research Ethics}, volume = {22}, number = {1}, pages = {21-37}, doi = {10.1177/17470161241309347}, abstract = {This study addresses the current lack of empirical data on the experiences and attitudes of clinical research professionals towards AI-powered clinical trial management tools. Clinical research professionals affiliated with various Swiss and international clinical research networks were invited to participate in an online survey. The survey focused on nine use cases of AI-powered clinical trial management tools. Participants were asked to share their ethical considerations, and their experiences were assessed at both the individual and institutional levels. Answers from 110 participants, primarily from Swiss academic institutions, were included in the data analysis. While participants generally held positive attitudes, mixed views were also common. Governmental attitudes were widely perceived as unclear or cautious, with the associated regulatory opacity possibly contributing to the limited adoption of AI in trial management observed in this study. In addition to a clear AI governance, ethical compliance and training emerged as crucial aspects. There was also broad consensus on the necessity of informing patients about tools’ uses and of ensuring an independent certification of the respective tools. The widespread adoption of AI-powered clinical trial management tools remains distant, entangled in a net of complexities that require further empirical investigation. Ethical concerns appear to play a crucial role, driven by distrust and the absence of a clear regulatory framework. Urgent actions are needed at both institutional and governmental levels, including the establishment of a governance framework and training provision. © The Author(s) 2025. This article is distributed under the terms of the Creative Commons Attribution-NonCommercial 4.0 License (https://creativecommons.org/licenses/by-nc/4.0/) which permits non-commercial use, reproduction and distribution of the work without further permission provided the original work is attributed as specified on the SAGE and Open Access pages (https://us.sagepub.com/en-us/nam/open-access-at-sage).}, keywords = {AI; Clinical research; clinical trial management; ethics; research ethics}, } @article{napolipmadis2026244, title = {On moving fast and breaking things.. again: social media’s lessons for generative AI governance}, author = {Napoli P.M. and Adi S.}, year = {2026}, journal = {Information Communication and Society}, volume = {29}, number = {6}, pages = {1912-1928}, doi = {10.1080/1369118X.2025.2513668}, abstract = {Generative AI systems are increasingly being employed globally, bringing with them both tremendous promise and substantial potential for harm. As is often the case, governance initiatives are lagging behind technological diffusion. This article looks to the very recent past–the rise of social media platforms and their associated algorithmic systems–as a guide for ongoing and future conversations about the governance of generative AI. Specifically, this article identifies a set of missteps that have characterized social media governance that have direct applicability to the generative AI context. In addressing this topic, this paper will explore issues such as the valorization of innovation as a guiding governance principle, the limitations of self-regulation, and the need for a dedicated regulatory body. All of these key components of the failure of social media governance have clear applications to, and lessons for, the governance of generative AI–a context in which the stakes may be even higher. © 2025 Informa UK Limited, trading as Taylor & Francis Group.}, keywords = {digital platforms; Generative AI; social media; technology policy}, } @article{schulzmloewencolnj2026245, title = {Preventing AI extractivism: the case for braiding indigenous data justice with ABS for stronger AI data governance}, author = {Schulz M. and Loewen-Colón J.}, year = {2026}, journal = {AI and Society}, doi = {10.1007/s00146-026-02931-z}, abstract = {Artificial-intelligence systems are rapidly reproducing colonial extractivism by harvesting Indigenous linguistic, biometric, geospatial, and ecological data without consent, compensation, or accountability. Biotechnology offers a blueprint for curbing such practices: the Convention on Biological Diversity and its Nagoya Protocol obligate users of genetic resources to obtain Prior Informed Consent, negotiate Mutually Agreed Terms, and share benefits fairly. No comparable framework restrains the digital appropriation that underpins many AI products. Consequently, corporations and states monetize Indigenous knowledge systems under the banners of “open data” and “scientific neutrality,” eroding rights affirmed in the United Nations Declaration on the Rights of Indigenous Peoples (UNDRIP). In response to this rising risk of AI extractivism, we make the case for a binding, sui generis ABS protocol for AI data governance. First, through a series of case studies we demonstrate that AI extraction mirrors the colonial and biopiracy controversies that originally triggered Access‑and‑Benefit‑Sharing (ABS) rules in biotechnology. Second, we translate those rules into a digital register by braiding two Indigenous data‑governance frameworks—OCAP® (Ownership, Control, Access, Possession) and the CARE Principles (Collective Benefit, Authority to Control, Responsibility, Ethics)—inside the ABS triad of consent, terms, and benefit‑sharing. The resulting model grounds technical safeguards in relational accountability and Indigenous legal orders. Such an instrument would compel transparent negotiations with Indigenous rights‑holders, assign enforceable authority over data across the AI lifecycle, and require equitable redistribution of the economic value generated by models trained on Indigenous data. Embedding ABS principles into AI governance offers a decolonial pathway that centers Indigenous epistemologies, promotes ethical foresight, and transforms AI from a vehicle of digital colonialism into a space for algorithmic justice. © The Author(s) 2026.}, keywords = {Access and benefit sharing; AI; Digital colonialism; Extractivism; Indigenous data justice; Indigenous data sovereignty}, } @article{vinjamurijk2026246, title = {From Charts to Chips: How Artificial Intelligence is Healing Healthcare}, author = {Vinjamuri J.K.}, year = {2026}, journal = {International Journal of Drug Delivery Technology}, volume = {16}, number = {6}, pages = {780-789}, doi = {10.25258/ijddt.16.6s.105}, abstract = {In healthcare, the shift is happening from old paper charts to innovative systems run by AI and large databases. This paper shows how AI is helping to modernize and improve patient care in the fields of diagnosis, analytical predictions, customized treatment and daily operations. AI is making it easier to identify health problems early and expect a decline in patients, able to support hospitals and their patients more effectively. It also covers the problems and difficulties involved in bringing AI into healthcare, including matters related to bias, privacy and the regulations in place. With medicine moving forward in this new digital era, working together with advanced technology should make care more accurate, efficient and caring. © 2026, Dr. Yashwant Research Labs Pvt. Ltd. All rights reserved.}, keywords = {AI Ethics; Artificial Intelligence; Diagnosis; Digital Transformation; Healthcare; Healthcare Technology; Personalized Medicine; Predictive Analytics}, } @article{tanveerugiahoangtishaqsdaij2026247, title = {Transformative Learning Through Organizational Learning Theory in the Gen-AI Landscape: Challenges and Capacities for Organizational Evolution}, author = {Tanveer U. and Gia Hoang T. and Ishaq S. and Dai J.}, year = {2026}, journal = {IEEE Transactions on Engineering Management}, volume = {73}, pages = {2939-2949}, doi = {10.1109/TEM.2026.3683981}, abstract = {Generative artificial intelligence (Gen-AI) marks a transformative phase for organizations, creating both major opportunities and challenges. Anchored in organisational learning theory, this study examines how organizations adopt Gen-AI by unpacking the related challenges, learning processes, and required capabilities. Using a multiple-case study design with semistructured interviews across four pioneering organizations, the research highlights the importance of adaptability, rapid decision-making, risk tolerance, and a strong learning culture in shaping organizational learning during Gen-AI adoption. The findings show that Gen-AI creates distinctive learning demands compared with earlier technologies. Its cognitive reasoning and adaptive functionalities require organizations to engage in iterative, continuous learning to integrate these tools into both strategic and operational systems. The study offers practical guidance for organizations seeking to overcome adoption challenges, strengthen learning cultures, and realize Gen-AI’s transformative potential. It also contributes to the broader Gen-AI discourse by providing a theoretically grounded and empirically supported roadmap for managing AI integration, including strategies such as iterative feedback loops, risk-tolerant decision-making, and embedding ethical considerations in AI governance. © 1988-2012 IEEE. All rights reserved.}, keywords = {Adaptive capacities; decision-making in AI adoption; generative artificial intelligence (Gen-AI); learning culture; learning processes; organisational learning theory (OLT); Artificial intelligence; Cognitive systems; Decision theory; Ethical technology; Federated learning; Iterative methods; Learning systems; Adaptive capacity; Decision-making in AI adoption; Decisions makings; Generative artificial intelligence; Learning cultures; Learning process; Learning Theory; Organizational learning; Organizational learning theory; Transformative learning; Decision making}, } @article{cucarinlaviolafespositodefalcos2026248, title = {Artificial intelligence in corporate governance domains: a scientometric and content analysis}, author = {Cucari N. and Laviola F. and Esposito De Falco S.}, year = {2026}, journal = {Corporate Governance (Bingley)}, pages = {1-41}, doi = {10.1108/CG-05-2024-0308}, abstract = {Purpose – This paper aims to examine the academic literature on artificial intelligence (AI) in corporate governance (CG), defined as research at the intersection of AI techniques and governance domains, encompassing applications, integration into board processes and regulatory or ethical implications. By combining bibliometric and content analysis, the study maps key strands of scholarship and outlines future research directions, thereby advancing the discourse on the role of AI in CG. Design/methodology/approach – The research uses bibliometric analyses using Bibliometrix and VOSviewer on a corpus of 122 academic papers from Scopus. The authors apply performance analysis and science mapping to scrutinise scholarly contributions and identify thematic trends. Further, manual content analysis of the 21 research papers was conducted. Findings – The findings demonstrate the transformative impact of AI on CG, fundamentally reshaping decision-making processes, operational efficiency, communication and diversity within board structures. While AI has initially been used primarily as an external support tool, future research may point to its potential role as an autonomous agent. This emerging influence is driving the evolution of CG practices, signalling a shift towards an era of “artificial corporate governance”. Originality/value – The study highlights the need for firms to understand the interplay between AI technologies and CG to navigate the changing landscape effectively. It provides original insights into the multifaceted impacts of AI on CG, beyond operational efficiency and data analytics and underscores the importance of further research in this field. © 2026 Emerald Publishing Limited}, keywords = {AI governance; AI-driven decision-making; Artificial governance; Artificial intelligence; Corporate governance; Literature review}, } @article{strohmeiersbeckermscheerwellere2026249, title = {Beyond aversion – principles of appropriate algorithmic decision-making in human resource management}, author = {Strohmeier S. and Becker M. and Scheer-Weller E.}, year = {2026}, journal = {Expert Systems with Applications}, volume = {296}, pages = {128954}, doi = {10.1016/j.eswa.2025.128954}, abstract = {As algorithmic decision-making (ADM) becomes increasingly embedded in human resource management (HRM), concerns such as a lack of fairness and accountability raise urgent questions about its appropriateness. This study addresses the need for ADM evaluation by developing a coherent framework of principles grounded in the task-technology fit approach. It elaborates a balanced triad of nine indispensable ADM principles—methodical (veracity, accuracy, validity), managerial (relevancy, quality, efficiency), and ethical (fairness, accountability, transparency)—and validates them through a systematic literature review of 126 ADM artifacts in HRM. The analysis reveals a troubling lack of attention to ethical and managerial dimensions, while even methodical aspects are often neglected—with the notable exception of accuracy. Building on these findings, the study outlines a forward-looking agenda to operationalize, calibrate, implement, evaluate, and codify ADM principles, ultimately promoting responsible, appropriate ADM in HRM that reflects an evaluative stance beyond mere aversion. © 2025 The Author(s)}, keywords = {AI Principles; Algorithmic Decision-Making; Ethical AI; Human Resource Management; Machine Learning; Prescriptive HR Analytics; Artificial intelligence; Behavioral research; Decision making; Delta modulation; Ethical technology; Learning systems; Natural resources management; Resource allocation; AI principle; Algorithmic decision-making; Algorithmics; Decision-making evaluation; Decisions makings; Ethical AI; Human resources management; Machine-learning; Management concerns; Prescriptive HR analytic; Human resource management}, } @article{kimdh2026250, title = {When trust becomes symbolic: standards and the governance of AI}, author = {Kim D.-H.}, year = {2026}, journal = {Information Technology and People}, pages = {1-20}, doi = {10.1108/ITP-01-2026-0128}, abstract = {Purpose – This paper aims to reconceptualise standardisation as a generative governance arrangement that organises expectations, trust and decision-making under persistent uncertainty. Drawing on Niklas Luhmann's systems theory and extending Kim's (2025a) concept of standards as thematic identities and expectational structures, it seeks to explain how standards operate beyond technical coordination. Focusing on the UK's AI standards strategy, the paper examines how standards and strategy function as programmes at semantic and structural levels. In doing so, it explores how AI assurance practices are shaped and trust recast as an organisationally produced outcome rather than an inherent property of AI systems. Design/methodology/approach – The study adopts a qualitative, theory-driven analytical approach grounded in systems theory. It combines conceptual reconstruction with an in-depth case analysis of the UK AI standards and assurance ecosystem. Primary materials include UK policy documents, standards strategies, assurance roadmaps and regulatory guidance, complemented by secondary academic literature on standardisation and AI governance. These materials are analysed through Luhmannian concepts such as second-order observation, decision premises and programmes, enabling a semantic and structural analysis of how standards are institutionalised and operationalised within contemporary AI governance. Findings – The paper finds that AI standards in the UK operate as multi-layered observational procedures that institutionalise trust by stabilising expectations rather than guaranteeing technological reliability. Standards and strategy function as interlinked programmes: semantically, they construct thematic identities such as “trusted AI”; structurally, they embed decision premises across assurance practices. The analysis shows that standards oscillate between dissemination media and symbolically generalised communication media, enabling coordination through simplified symbols like certification. This oscillation produces ambivalent effects, simultaneously facilitating governance while fostering strategic ignorance and the ritualisation of AI ethics. Originality/value – This paper makes an original contribution by systematically linking Niklas Luhmann's core concepts – second-order observation, decision premises and programmes – to the standards and standardisation literature. Extending Kim's (2025a) concept of standards as thematic identities and expectational structures, it reconceptualises AI assurance as a multi-layered observational procedure in which trust is produced through recursive processes of decision and validation rather than embedded in AI systems. By theorising standards as mediating forms that oscillate between dissemination media and symbolically generalised communication media, the paper offers a novel Luhmannian explanation of how standards stabilise expectations while generating ambivalence, strategic ignorance and the ritualisation of ethics in contemporary AI governance. © Emerald Publishing Limited}, keywords = {Government policy; IT governance; Organizational theory; Socio-technical theory; Standard; Technology; Trust}, } @article{peppledmuthuthantrigen2026251, title = {Artificial intelligence, innovation and the new architecture of exploitation: Towards reconfiguring humanness in the age of algorithmic labour}, author = {Pepple D. and Muthuthantrige N.}, year = {2026}, journal = {Journal of Innovation and Knowledge}, volume = {11}, pages = {100878}, doi = {10.1016/j.jik.2025.100878}, abstract = {Purpose This conceptual study explores how artificial intelligence (AI) is transforming the nature of work and reconfiguring the experience of humanness, particularly among low-skilled and informal workers. Method Using an integrative literature review methodology, the study synthesises interdisciplinary research from organisational studies, sociology, and AI ethics to examine the mechanisms through which AI-driven labour displacement, algorithmic management, and structural precarity contribute to new forms of exploitation. Findings The study develops a novel conceptual framework that links technological transformation to the erosion of the relational, moral, and emotional dimensions of work conditions, resulting in conditions increasingly resembling modern slavery. Originality the study’s novelty lies in its reframing of AI as a socio-technical actor with ontological consequences for worker identity, autonomy, and dignity. The findings underscore the need for ethical AI design, inclusive policy frameworks, and human-centred organisational practices. Practical implications This paper offers practical implications for policymakers, technologists, and business leaders seeking to align innovation with social justice and sustainable labour futures. Plain summary Artificial intelligence (AI) is reshaping the nature of work and disrupting the human experience, especially for low-skilled and informal workers, highlighting the urgency and complexity of this research. AI-driven labour displacement and algorithmic management contribute to new forms of exploitation that echo modern slavery. The erosion of humanness at work is linked to reduced autonomy, empathy, and moral agency under opaque algorithmic systems. A socio-technical framework is needed to address AI’s impact on dignity and agency, with ethical design and inclusive governance at its core. JEL Code O330, O31, O32 © 2025 The Author(s).}, keywords = {Artificial intelligence; Digital labour; Ethical innovation; Humanness; Labour displacement; lgorithmic management; Modern slavery}, } @article{liuzyrwangytyanjjguptasgiannakism2026252, title = {Exploring the Relationship Between Human-Centric AI and firm Idiosyncratic Risks}, author = {Liu Z.-Y.R. and Wang Y.-T. and Yan J.-J. and Gupta S. and Giannakis M.}, year = {2026}, journal = {Information Systems Frontiers}, doi = {10.1007/s10796-026-10759-7}, abstract = {Despite the extensive discussions of human-centric AI (HCAI) in Industry 5.0, its effects on firms’ idiosyncratic risks (IR) remains underexplored. This is an imperative issue for firms navigate financial risks during the current technological revolution, as IR reflects investor reactions to corporate heterogeneous AI strategies and implementations by isolating firm-level stock volatility from systematic factors. Integrating situated AI theory with social-technical systems theory, we conceptualise HCAI as a situated AI strategy that reduces AI-related ethical risks and fosters AI-Human synergies in firms’ business operations by aligning with stakeholders’ diverse expectations. Moreover, socio-technical factors, namely digitalisation, operational efficiency, executive shareholding, and CEOs with IT background, may moderate the HCAI-IR relationship. Using a multi-source panel dataset of Chinese listed firms from 2015 to 2023, we find that HCAI is associated with lower firm IR. Furthermore, digitalisation and executive shareholding strengthen this risk-reducing effect, whereas operational efficiency and CEOs with IT background surprisingly attenuate it. Our findings offer theoretical contributions and practical insights for both ethical AI governance and firm financial risk management in the AI era. © The Author(s) 2026.}, keywords = {Human-centric AI; Idiosyncratic risk; Large language models; Situating AI theory; Socio-technical systems theory; Artificial intelligence; Economic and social effects; Ethical aspects; Investments; System theory; Financial risks; Human-centric; Human-centric AI; Idiosyncratic risk; Language model; Large language model; Operational efficiencies; Shareholdings; Situating AI theory; Sociotechnical systems theory; Risk management}, } @article{liumliylij2026253, title = {The impact of generative AI use on employees’ psychological distress: a moderated mediation model}, author = {Liu M. and Li Y. and Li J.}, year = {2026}, journal = {Frontiers in Public Health}, volume = {14}, pages = {1798423}, doi = {10.3389/fpubh.2026.1798423}, abstract = {Introduction – With the widespread adoption of generative artificial intelligence (Gen AI) in the workplace, concerns have emerged about whether employees experience psychological distress alongside efficiency improvements. Methods – Based on the Stimulus-Organism-Response (SOR) framework, we developed a moderated mediation model and used survey data from 424 Chinese employees to examine the relationship between Gen AI use and employees’ psychological distress. We tested dual mediating pathways (job insecurity and workplace loneliness) and two boundary conditions (information literacy and AI ethical risk perception). Results – Our study reveals that Gen AI use is significantly associated with psychological distress via the dual mediating paths of job insecurity and workplace loneliness. Employees’ information literacy can mitigate the relationship between Gen AI and job insecurity, and further moderate the mediating effect on psychological distress. Conversely, AI ethical risk perception strengthens the relationship between Gen AI and workplace loneliness, and is further associated with psychological distress via a moderated mediation effect. Discussion – Our study contributes to the theoretical understanding and empirical evidence regarding the relationship between Gen AI and employees’ psychological distress. It also provides practical recommendations for managers and employees on how to interact effectively with Gen AI tools. Copyright © 2026 Liu, Li and Li.}, keywords = {AI ethical risk perception; Gen AI; information literacy; job insecurity; psychological distress; workplace loneliness; Adult; China; Female; Generative Artificial Intelligence; Humans; Job Security; Loneliness; Male; Mediation Analysis; Middle Aged; Psychological Distress; Stress, Psychological; Surveys and Questionnaires; Workplace; adult; China; distress syndrome; female; generative artificial intelligence; human; job security; loneliness; male; mediation analysis; mental stress; middle aged; psychology; questionnaire; workplace}, } @article{pratonoahelgekahwsjeongbg2026254, title = {Does being literate in AI make workplaces more equal? The mediating role of psychological capital}, author = {Pratono A.H. and Elgeka H.W.S. and Jeong B.G.}, year = {2026}, journal = {International Journal of Sociology and Social Policy}, pages = {1-17}, doi = {10.1108/IJSSP-02-2026-0125}, abstract = {Purpose – This study examines how AI literacy shapes workplace equality and competitive advantage, with particular attention to the mediating role of psychological capital. Design/methodology/approach – Using survey data from 467 owner-managers, managers and senior staff in small- and medium-sized enterprises in Surabaya, Indonesia, the study applies partial least squares structural equation modeling (PLS-SEM) to test the proposed relationships. Findings – AI literacy exhibits a statistically significant yet modest direct effect (ß = 0.133, p < 0.05) on workplace equality and competitive advantage. However, its primary influence operates indirectly through psychological capital (strong path: ß = 0.600, p < 0.001), which fosters self-efficacy, resilience, optimism and hope, thereby driving equality (ß = 0.531, p < 0.001) and firm performance. Psychological capital, in turn, significantly promotes workplace equality, emerging as a pivotal antecedent to competitive advantage (ß = 0.494, p < 0.001). These patterns underscore AI literacy's role as a human-centered amplifier in resource-constrained SMEs. Practical implications – Beyond technical training, integrate PsyCap-building interventions (e.g. resilience workshops paired with AI ethics modules) into SME support programs, drawing on evidence from Indonesian contexts to foster inclusive AI adoption and align with UNESCO equity goals. Originality/value – By conceptualizing AI literacy as a socially embedded capability, this study demonstrates that AI functions as a social amplifier: its performance and equality outcomes depend less on technology itself than on how it reshapes human agency and psychological capacity in organizational contexts. © Emerald Publishing Limited}, keywords = {AI literacy; Competitive advantage; Equality; Psychological capital}, } @article{chentgascohernandezmgilgarciajr2026255, title = {Understanding Accountability in AI Use Within the Public Sector: Insights From State Governments in the United States}, author = {Chen T. and Gasco-Hernandez M. and Gil-Garcia J.R.}, year = {2026}, journal = {Public Administration}, doi = {10.1111/padm.70051}, abstract = {Recent research has examined various aspects of the government's use of artificial intelligence (AI), including its implications for accountability as well as the policy and management recommendations proposed to address some related concerns. However, two important gaps remain and need further empirical exploration. First, it is unclear why accountability in AI-based system use is viewed as such a critical issue. Second, it is uncertain to what extent existing policy and management recommendations from academia are reflected in government policies. This study aims to bridge these gaps by empirically investigating how the public sector addresses accountability in the use of AI-based systems. Based on an analysis of policy documents related to AI-based systems from 32 U.S. states, our findings show why accountability matters, who is being held accountable to whom, by what standards, and with what consequences from a practical perspective. Furthermore, we identify several research-practice gaps that merit further exploration. © 2026 John Wiley & Sons Ltd.}, keywords = {accountability; AI governance; artificial intelligence; generative AI; transparency}, } @article{albayraktaroluamergenagven2026256, title = {Generative AI Models as Wicked Resources: A Dynamic Perspective on Resource Governance}, author = {Albayraktaroğlu A. and Mergen A. and Güven Ç.}, year = {2026}, journal = {Journal of Management}, doi = {10.1177/01492063261434193}, abstract = {The proliferation of generative AI models fundamentally alters organizational capabilities, enabling novel value creation while challenging incumbent governance frameworks. Employing a phenomenon-driven approach, this study integrates and extends property rights theory (PRT) and stakeholder resource-based theory (SRBT) to address governance challenges posed by generative AI. By leveraging system dynamics modeling, we conceptualize the dynamic interplay among stakeholder claims, institutional arrangements, and value appropriation outcomes, highlighting how feedback loops, delays, and accumulations shape these interactions. Our analysis reveals two insights: First, stable equilibrium states in stakeholder claims and property rights arrangements may not invariably lead to equitable outcomes, due to stakeholder power disparities and attribution ambiguity associated with generative AI. Second, framing the evolution of generative AI models as organizational resources from the complementary perspectives of PRT and SRBT reveals distinct resource features largely unexamined in the strategy literature. Hence, we introduce the concept of “wicked resources,” characterizing generative AI models by their inherent attribution ambiguity and emergent unpredictability. Building on prior research on resource complexity and uncertainty in the strategy literature, wicked resources are marked by the difficulty firms face in delineating and enforcing control within shifting sociopolitical contexts. This paper makes three key contributions: addressing the dynamic, multi-stakeholder nature of generative AI governance; introducing wicked resources as a novel resource category in strategy and management literatures; and identifying theoretical gaps, advocating for a dynamic, systemic approach to property rights and stakeholder bargaining. © The Author(s) 2026. This article is distributed under the terms of the Creative Commons Attribution-NonCommercial 4.0 License (https://creativecommons.org/licenses/by-nc/4.0/) which permits non-commercial use, reproduction and distribution of the work without further permission provided the original work is attributed as specified on the SAGE and Open Access pages (https://us.sagepub.com/en-us/nam/open-access-at-sage).}, keywords = {generative AI; property rights theory; resource governance; stakeholder resource-based theory; wicked resources}, } @article{jiangqzhumxianxguoxlius2026257, title = {Debate: The efficiency-accountability dilemma—transparency challenges in digital government AI governance}, author = {Jiang Q. and Zhu M. and Xian X. and Guo X. and Liu S.}, year = {2026}, journal = {Public Money and Management}, volume = {46}, number = {5}, pages = {540-541}, doi = {10.1080/09540962.2026.2614325}, abstract = {[No abstract available]}, } @article{leonm2026258, title = {Framework for the Ethical and Societally Conscious AI Deployment}, author = {Leon M.}, year = {2026}, journal = {WSEAS Transactions on Information Science and Applications}, volume = {23}, pages = {262-275}, doi = {10.37394/23209.2026.23.20}, abstract = {This paper presents the Algorithmic Impact Navigator (AIN), a structured governance framework for the responsible development and deployment of Artificial Intelligence (AI) systems. AIN offers a systematic, lifecycle-oriented methodology for identifying benefits, anticipating ethical and legal risks, and evaluating societal impact. It embeds ethical reasoning into design, implementation, and post-deployment monitoring, emphasizing reliability, safety, transparency, and human-centered values. The framework supports structured stakeholder engagement, explicit trade-off analysis, and rigorous documentation to enhance accountability and auditability. By aligning ethical principles with organizational strategy and regulatory obligations, AIN enables informed decision-making across multidisciplinary teams. The paper argues that ethical AI governance is not antithetical to innovation but rather foundational to sustainable and trustworthy technological progress. As AI systems increasingly influence social, economic, and institutional outcomes, frameworks such as AIN are essential for safeguarding individual rights, maintaining public trust, and ensuring long-term societal value. © 2026 World Scientific and Engineering Academy and Society. All rights reserved.}, keywords = {AI lifecycle accountability; Algorithmic impact assessment; Ethical AI governance; Human-centered AI; Responsible AI deployment; Transparency and explainability in AI}, } @article{subc2026259, title = {AI-Powered Product Placement: A New Framework for Computational Persuasion}, author = {Su B.-C.}, year = {2026}, journal = {International Journal of Human-Computer Interaction}, doi = {10.1080/10447318.2026.2659304}, abstract = {Artificial intelligence transforms product placement from static insertion into dynamic computational persuasion within streaming media. This study analyzes technological architectures and marketing efficacy through comparative case studies at Netflix, Amazon, and Google. We address the gap between static media theories and recursive artificial intelligence by introducing the Algorithmic Awareness construct. This construct explains non linear viewer resistance to hyper personalization. The research utilizes a hierarchical structure reflecting nested learning processes distributed across multiple time scales. Empirical results substantiated by T tests and Analysis of Variance demonstrate that artificial intelligence driven placements significantly improve brand recall and Return on Investment. However, these capabilities introduce critical risks concerning data privacy and manipulation. We propose three integrated theoretical frameworks modeling technological, marketing, and ethical dimensions of this phenomenon. This study establishes a foundational governance model for interplay of artificial intelligence and advertising while offering scholarly and managerial imperatives for digital ecosystems. © 2026 Taylor & Francis Group, LLC.}, keywords = {AI ethics; Artificial intelligence; computational persuasion; product placement; streaming media; Commerce; Computation theory; Data privacy; Ethical aspects; Investments; Marketing; Media streaming; AI ethic; Algorithmics; Case-studies; Computational persuasion; Google+; Medium theory; Netflix; Product placement; Streaming medium; Technological architectures; Artificial intelligence}, } @article{sposatomdittmarecportillojpv2026260, title = {Navigating the dark side of AI in service ecosystems: an ethical leadership framework for risk mitigation; [穿越服务生态系统中人工智能的阴暗面：基于伦理领导力的风险缓解框架]}, author = {Sposato M. and Dittmar E.C. and Portillo J.P.V.}, year = {2026}, journal = {Service Industries Journal}, doi = {10.1080/02642069.2026.2643384}, abstract = {The rapid integration of artificial intelligence (AI) into service ecosystems is transforming value cocreation while generating significant ethical risks that threaten customer trust, organisational legitimacy, and social sustainability. This paper develops the Ethical AI Risk Mitigation (EAIRM) model to examine how different configurations of human-AI collaboration create distinct ethical challenges across fairness, autonomy, transparency, and accountability dimensions. Drawing on a structured literature synthesis, we identify four leadership approaches (compliance-oriented, values-based, stakeholder-engaged, and anticipatory) that systematically mitigate ethical risks while enabling service innovation. Through integrative theory building, the model contributes to service research and practice by: (1) revealing how identical ethical risks operate through different causal mechanisms depending on human-AI resource configuration; (2) specifying multi-actor governance structures for service ecosystems where no single actor controls ethical outcomes; (3) theorizing leadership mechanisms and organisational mediators that convert ethical principles into operational practices; and (4) generating testable propositions with boundary conditions, moderators, and feedback dynamics. This framework advances service ecosystem theory by demonstrating that resource relations carry ethical risk implications requiring polycentric governance, not merely value creation potential. © 2026 Informa UK Limited, trading as Taylor & Francis Group.}, keywords = {AI governance; Artificial intelligence; ethical leadership; human-AI relations; responsible AI; risk mitigation; service ecosystems}, } @article{canatomc2026261, title = {From ethics washing to enforceable accountability: hospitals as hubs of digital constitutionalism}, author = {Canato M.C.}, year = {2026}, journal = {AI and Society}, doi = {10.1007/s00146-026-02980-4}, abstract = {The governance of artificial intelligence (AI) in European healthcare reveals a structural paradox. While the EU Artificial Intelligence Act (AI Act) classifies most medical AI systems as high-risk and imposes extensive obligations on providers and deployers, the institutional mechanisms meant to enforce these obligations often remain merely advisory. This article conceptualizes this disjunction as the enforceability gap: the systemic mismatch between formal legal responsibility and the institutional capacity to exercise coercive authority. This paper, in bridging insights from accountability theory, risk governance, and digital constitutionalism, argues that enforceability constitutes the missing link between algorithmic authority and democratic legitimacy. Authority without enforceability, it contends, results in symbolic governance, where ethics boards simulate oversight without power to constrain decisions. To overcome this deficit, the paper proposes a model of enforceable accountability grounded in four pillars: legally empowered institutional boards with conditional veto powers; corporate and professional liability mechanisms; transparency and restorative-oriented frameworks; and participatory procedures embedding democratic deliberation. The paper’s key innovation is to re-center hospitals as one-stop accountability hubs—integrated institutional loci where ethical review, legal compliance, and technical audit converge within a unified governance structure. By conceptualizing hospitals as active co-regulators rather than passive deployers, the paper demonstrates how enforceable accountability can be operationalized at the point of clinical implementation. In this perspective, addressing the enforceability gap is not solely a matter of regulatory refinement, but a structural condition for aligning high-risk AI governance with the rule-of-law requirements of the European digital order, ensuring that ethical obligations acquire effective legal force rather than remaining aspirational norms. © The Author(s) 2026.}, keywords = {Algorithmic Accountability; Artificial Intelligence Governance; Digital Constitutionalism; Enforceability Gap; EU Artificial Intelligence Act; Healthcare AI; Artificial intelligence; Ethical technology; Government data processing; Hospitals; Product liability; Algorithmic accountability; Algorithmics; Artificial intelligence governance; Artificial intelligence systems; Digital constitutionalism; Enforceability gap; EU artificial intelligence act; Healthcare artificial intelligence; Institutional capacities; Power; Health care}, } @article{succchannk2026262, title = {Assembling platform governance as private ordering in the age of generative AI: platform interdependence in policy evolution}, author = {Su C.C. and Chan N.K.}, year = {2026}, journal = {Information Communication and Society}, volume = {29}, number = {6}, pages = {1929-1953}, doi = {10.1080/1369118X.2025.2513672}, abstract = {The rise of generative AI, such as ChatGPT, has transformed platform governance, affecting not only their own frameworks but also those of interconnected platforms like Twitter/X, TikTok, and Facebook. This study explores the evolution, interdependence, and assembling of platform governance through a comparative analysis of policy documents, including Community Guidelines, Privacy Policies, and Terms of Service. By examining OpenAI’s policies alongside those of major social media platforms from 2022 to 2024, we trace the co-evolution of platform values such as privacy, engagement, and accountability. Using longitudinal lexical and time-series analyses, we identify three prominent value patterns: positively-aligned values, divergent values, and floating values. These findings suggest that OpenAI’s policy have a significant influence on other platforms, particularly in aligning privacy and accountability while revealing divergences in values such as user choice and platform power. Drawing on assemblage thinking, we argue that governance is an ongoing process contingent on the interplay between heterogeneous entities, where platforms engage in private ordering to shape and legitimize their governance structures. The study highlights the interconnectedness of platform governance and the complex ways in which generative AI reshapes policy frameworks across the digital ecosystem. We conclude by reflecting on the implications for both platforms and policymakers, emphasizing the need for coordinated regulation in the face of evolving AI governance. © 2025 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group.}, keywords = {assemblage; generative AI; interdependence; Platform governance; policy}, } @article{monteiroa2026263, title = {The evolving mandate of project management offices: governance, innovation, and performance, evidence from a longitudinal case study}, author = {Monteiro A.}, year = {2026}, journal = {Business Process Management Journal}, pages = {1-22}, doi = {10.1108/BPMJ-09-2025-1464}, abstract = {Purpose – This study investigates the long-term evolution of the Project Management Office (PMO) within the foundation of Organizational Project Management (OPM). It examines how the PMO'smandate, authority, and responsibilities developed in a large European organization, tracing its transition from departmental support to strategic governance and, more recently, to a central role in innovation and artificial intelligence (AI) initiatives. Design/methodology/approach – A longitudinal single-case study design was employed, combining semi-structured interviews, internal documentation, and a follow-up questionnaire with Enterprise PMO (EPMO) professionals to identify PMO functions across stages and analyze mandate and authority expansion. Findings – Three main findings emerged. First, the PMO evolved through four stages, expanding from IT support to enterprise-wide governance. This followed a layered mandate logic, where foundational functions were retained while governance and strategic roles were progressively added. Second, a decisive transformation occurred when the PMO was relocated from IT to board-level reporting, expanding its autonomy, process performance, and strategic influence. Third, the PMO was explicitly mandated to lead innovation and AI initiatives, positioning it as both a governance body and a driver of digital transformation. Originality/value – This study provides a rare longitudinal examination of PMO evolution in the context of OPM, addressing limited empirical evidence on sustained PMO development beyond early support and governance roles. It advances the concept of the layered mandate and offers guidance for structuring PMOs to sustain stability, expand authority, and assume responsibility for innovation and AI governance. © Emerald Publishing Limited}, keywords = {Innovation; Longitudinal study; Organizational performance; Project governance; Project management; Project management office}, } @article{bektihpancasilawanrkomarasrsofiaturrohmahs2026264, title = {Lessons learned from global practices and public leadership toward mature AI regulation for e-government in Indonesia}, author = {Bekti H. and Pancasilawan R. and Komara S.R. and Sofiaturrohmah S.}, year = {2026}, journal = {Cogent Social Sciences}, volume = {12}, number = {1}, pages = {2652002}, doi = {10.1080/23311886.2026.2652002}, abstract = {The rapid adoption of artificial intelligence (AI) in e-government presents both opportunities and significant ethical and regulatory challenges, particularly in ensuring transparency, accountability, and public trust. This study aims to analyze the current state of AI regulation in Indonesia’s e-government and to identify lessons learned from advanced AI governance arrangements. Employing an exploratory qualitative case study design, this research relies on a qualitative synthesis of secondary data sources with data triangulation used to enhance analytical rigor. The findings indicate that Indonesia’s AI governance in e-government remains underdeveloped, characterized by weak regulatory coordination and the absence of explicit standards for transparency, accountability, and data protection, which poses risks to ethical governance and public trust. In contrast, Japan, Singapore, and the European Union have developed more mature and adaptive AI governance frameworks that emphasize human-centric principles, legal clarity, and institutional oversight. Drawing on lesson-drawing as a conceptual and comparative analytical framework, this study argues that Indonesia can strengthen its AI regulation in e-government by selectively adapting regulatory principles rather than replicating institutional models. Such an approach can support the development of contextually appropriate AI governance that aligns technological innovation with ethical standards, citizen protection, and trustworthy digital governance. © 2026 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group.}, keywords = {AI regulations; e-government; ethical AI principles; lessons learned; regulatory frameworks}, } @article{embongahkhairuldinwmkfwhassansarahmanahamutalibnaismaililm2026265, title = {MADANI PLANNING IN THE ERA OF AI: STRATEGIES FOR BALANCING TECHNOLOGICAL INNOVATION AND COMMUNITY WELL-BEING}, author = {Embong A.H. and Khairuldin W.M.K.F.W. and Hassan S.A. and Rahman A.H.A. and Mutalib N.A. and Ismail I.L.M.}, year = {2026}, journal = {Planning Malaysia}, volume = {24}, number = {2}, pages = {63-79}, doi = {10.21837/PM.V24I41.1987}, abstract = {This paper explores the intersection of artificial intelligence (AI) and Malaysia Madani, a governance framework that prioritizes sustainability, prosperity, innovation, respect, trust, and compassion. The study employs a qualitative, exploratory design using inductive thematic document analysis of recent Malaysia Madani policy frameworks, official government reports, and comparative academic literature. Three core strategies are identified for balancing technological innovation with community well-being: the development of multidimensional well-being indicators, the adoption of community-based approaches in AI governance, and the design of ethical interventions that safeguard dignity and social cohesion. Definitions of Malaysia Madani planning, well-being, and AI are outlined, followed by an assessment of threats and opportunities posed by AI to social equity, cultural diversity, and democratic trust, with specific attention to marginalized communities in Malaysia. The findings suggest that embedding Malaysia Madani values into AI governance offers a unique model for aligning innovation with normative commitments to justice, inclusivity, and resilience. The paper concludes by calling for interdisciplinary, transdisciplinary, and systems-based approaches to ensure that AI becomes a tool for collective flourishing rather than a source of fragmentation. © 2026 by MIP}, keywords = {Artificial Intelligence (AI); Digital Governance; Malaysia Madani Values; Policy Frameworks; Sustainable Urban Planning}, } @article{sienickimsienickik2026266, title = {The Algorithmic Arbitrator: From the New York Convention to AI Rule of Law, Governance, and Enforceability of AI Use in Arbitration}, author = {Sienicki M. and Sienicki K.}, year = {2026}, journal = {Journal of International Arbitration}, volume = {43}, number = {2}, pages = {211-246}, doi = {10.54648/joia2026009}, abstract = {This article proposes a two-track framework for using artificial intelligence in international arbitration while preserving enforceability under the New York Convention (NY Convention) and due-process minimums. Track one, Artificial Inteligence (AI)-assisted arbitration, keeps human arbitrators fully responsible for fact-finding, legal reasoning, and the signed award, while using AI for document handling, translation, retrieval, and drafting under disclosure, symmetric access, and strict version control. Track two, AI-exclusive arbitration, treats a certified AI pipeline as the merits decision-maker and is recommended only for narrowly scoped sandbox pilots (highly structured, low-value, high-volume disputes) with explicit consent, frozen configurations, integrity logging, and a human legality/due-process backstop. To help courts apply existing refusal grounds without reopening the merits, we introduce a proportional AI Usage and Provenance Dossier (tool/version disclosure, hash manifests, sealed logs, exception reporting, and explicit explainability limits). We connect these operational controls to EU compliance anchors (General Data Protection Regulation (GDPR) and the EU AI Act) and to practical threats (prompt injection, retrieval poisoning, drift, and log omission), emphasizing that cryptographic artifacts provide tamper-evidence for recorded steps, not guarantees of correctness or completeness. © 2026 Kluwer Law International BV, The Netherlands}, keywords = {AI governance; AI-assisted arbitration; due process; enforceability; EU AI Act; GDPR; international arbitration; New York Convention; provenance logging; UNCITRAL Model Law}, } @article{idrismychalkmb2026267, title = {Trust by design: Crossing the chasm between clinical AI/ML innovation and practice}, author = {Idris M.Y. and Chalk M.B.}, year = {2026}, journal = {Journal of the National Medical Association}, doi = {10.1016/j.jnma.2026.03.008}, abstract = {Clinical artificial intelligence (AI) and machine learning (ML) have advanced rapidly, yet few systems achieve sustained use in routine clinical care. This innovation–practice gap reflects not a lack of model performance, but persistent failures in validation, governance, and trust when AI is deployed in real-world clinical environments. Models that perform well retrospectively often degrade prospectively, misalign with local workflows, or are ultimately abandoned, with current responses relying on people- and process-centric mechanisms such as expert review, bespoke validation, and discretionary governance. These approaches are slow, fragile, and difficult to scale, making trust the primary bottleneck to adoption. We argue for a shift toward trust by design, in which trust is treated as a system property rather than a judgment conferred by individuals or committees. We describe how trusted systems, including privacy-enhancing compute, trusted execution environments, and trusted research environments, embed enforceable guarantees around data use, auditability, and reproducibility. By operationalizing trust through infrastructure, these systems enable scalable local validation, reduce duplicated effort, and support more equitable and durable deployment of clinical AI/ML. © 2026 The Authors. Published by Elsevier Inc. on behalf of National Medical Association. This is an open access article under the CC BY-NC-ND license. http://creativecommons.org/licenses/by-nc-nd/4.0/}, keywords = {Clinical AI governance; Clinical artificial intelligence; Machine learning; Trust by design; Trusted systems; article; artificial intelligence; controlled study; decision making; human; infrastructure; machine learning; medical society; open access publishing; practice gap; privacy; reproducibility; trust; workflow}, } @article{oxusggagnonsfitsilisf2026268, title = {Intégration de l'intelligence artificielle dans les parlements francophones}, author = {Oxéus G. and Gagnon S. and Fitsilis F.}, year = {2026}, journal = {Parliaments, Estates and Representation}, doi = {10.1080/02606755.2026.2684191}, abstract = {This article examines the strategic integration of artificial intelligence (AI) within Francophone parliaments, an area that remains overlooked in current research on digital governance. While AI is increasingly used to support legislative functions such as transcription, translation, and document processing, Francophone institutions encounter structural challenges stemming from linguistic biases inherent in English-centric Large Language Models (LLMs) and the scarcity of high-quality French-language data. These limitations pose specific risks to multilingual and culturally diverse political systems. To address these issues, the study explores two main questions: How do Francophone parliaments’ AI strategies differ from those of other linguistic groups? In what ways do the French language and specific political cultures influence AI adoption? Utilizing a structured expert survey based on a high-level strategic framework covering strategy, implementation, and governance, this study analyzes data from parliamentary experts across diverse regions. Drawing on data collected, the results indicate extensive experimentation with AI but reveal fragmented implementation, minimal formal governance structures, and a reliance on quick-win applications, that is, on short-term solutions. Thus, the findings underscore the necessity for culturally and linguistically tailored AI tools that adhere to regional legal and normative frameworks. They also show that a more strategic approach to studying and deploying AI can lead to a more integrated framework, offering numerous opportunities for institutional development. This research provides the first systematic evaluation of AI adoption among Francophone parliaments and delineates key priorities to inform an updated research agenda, thereby supporting coherent and sustainable digital transformation across the Francophonie. © 2026 International Commission for the History of Representative and Parliamentary Institutions/Commission Internationale pour l’Histoire des Assemblées d’ États.}, keywords = {AI governance structures; digital transformation; Francophone parliaments; large language models (LLMs); multilingual data challenges; strategic AI integration}, } @article{aramburumgarcajgainesstouleimatnlamattinaeakhgarbkavallierosd2026269, title = {From prototype to deployment: An EU-centric lifecycle framework for law enforcement AI}, author = {Aramburu M. and García J. and Gaines S. and Touleimat N. and La Mattina E. and Akhgar B. and Kavallieros D.}, year = {2026}, journal = {Open Research Europe}, volume = {6}, pages = {141}, doi = {10.12688/openreseurope.21859.1}, abstract = {Artificial intelligence can substantially enhance law-enforcement capabilities, but its use in security research domains, including Fight Crime Terrorism, Border Management, INFRA, and DRS, raises significant legal, ethical, and operational challenges. Access to operational case data is typically restricted, making it unavailable for continuous ingestion or model training, while the adoption of third-party models, datasets and software artefacts introduces intellectual-property and licensing constraints. At the same time, EU regulations, notably the GDPR, the Law Enforcement Directive (LED), and the EU AI Act, impose procedural and technical safeguards that must be embedded throughout the development lifecycle. To address these challenges, this paper presents a practical, EU-centric lifecycle framework for developing AI systems in security-sensitive contexts. The methodology is structured into five stages: Matchmaking, Definition & Design, Development, Validation, and Monitoring. By mapping legal and ethical obligations to concrete engineering checkpoints, the framework supports data provenance, reproducibility, and software supply-chain assurance through artefacts such as dataset registries, Model Cards, and SBOMs. To address restricted access to operational data, the methodology also defines validation patterns for end-user evaluation, including on-premises bring-solution-to-data assessment. The main contributions of the paper are a tailored lifecycle methodology, a compliance mapping linking EU obligations to lifecycle evidence, and a practical assurance package for traceable and auditable development. The methodology is further illustrated through a worked example derived from the STARLIGHT (https://starlight-h2020.eu/) European project, showing how operational validation can be conducted without exposing raw law-enforcement data. Copyright: © 2026 Aramburu M et al.}, keywords = {Law Enforcement; Trustworthy AI; MLOps; DevSecOps; EU AI Act; GDPR; AI Governance; Security Research}, } @article{obnascaapanaierivag2026270, title = {Holistic Semantic Sustainability: Redesigning Organizational Sense-Making for AI Ethics Professions}, author = {Obnasca A. and Panai E. and Riva G.}, year = {2026}, journal = {Cyberpsychology, Behavior, and Social Networking}, pages = {21522715261452791}, doi = {10.1177/21522715261452791}, abstract = {[No abstract available]}, keywords = {article; human; occupation}, } @article{chencweimzhangpluopmengl2026271, title = {An infographic framework of GeoAI ethics based on news data}, author = {Chen C. and Wei M. and Zhang P. and Luo P. and Meng L.}, year = {2026}, journal = {Cartography and Geographic Information Science}, volume = {53}, number = {2}, pages = {188-208}, doi = {10.1080/15230406.2025.2593949}, abstract = {Artificial intelligence (AI) is widely used in geographic information science (GIS) and has spawned a new research direction, Geospatial Artificial Intelligence (GeoAI). While there is a growing awareness of potential negative impacts of AI and increased research on AI ethics, systematic ethics research on GeoAI is still lacking and needs the integration of quantitative analytical capabilities. However, existing research on GeoAI ethics lacks a methodological approach for deriving quantitative analytical results, which in turn limits the ability to obtain conclusions with numerical significance. To address this challenge, we collect news data to obtain raw cases of GeoAI ethics. Subsequently, an exploratory framing serves as the guideline for data coding, resulting in a recoded and quantifiable dataset, which is intended for conducting an exploratory analysis of the themes and content of GeoAI ethical issues as they appear in news data. By integrating the designed icon system, we conduct a visual analysis of the dataset. Based on the findings, an infographic framework for GeoAI ethics is proposed to strengthen the structured development of ethical GeoAI. The exploratory framing, together with the infographic framework, contributes to the systematic identification of ethical issues and the formulation of policy guide lines within the GeoAI community. © 2025 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group.}, keywords = {AI ethics; exploratory framing; GeoAI; infographics; visual analysis; Data handling; Ethical technology; Geographic information systems; Analytical results; Artificial intelligence ethic; Ethical issues; Exploratory framing; Geo-spatial; Geographic information science; Geospatial artificial intelligence; Infographic; Methodological approach; Visual analysis; analytical framework; artificial intelligence; ethics; GIS; quantitative analysis; visual analysis; Artificial intelligence}, } @article{varshneydvarshneya2026272, title = {Exploring AI ethical considerations and governance in the financial sector: examination of select countries}, author = {Varshney D. and Varshney A.}, year = {2026}, journal = {Journal of Modelling in Management}, pages = {1-30}, doi = {10.1108/JM2-09-2025-0467}, abstract = {Purpose – This paper aims to examine artificial intelligence (AI)-related ethical considerations and governance in select countries through an exploratory approach, studying recent research in the domain. Design/methodology/approach – A model with an appropriate rationale has been developed to address the gaps in AI ethical considerations and governance, complementing the prevailing shortcomings of AI systems and processes in the financial sectors of select countries. Findings – The authors found that some countries had stringent AI considerations and guidelines, while others had comparatively loose guidelines. To address the inconsistencies, a model covering the relevant dimensions has been framed, which can fill the current gaps in the countries studied or to be analyzed using this framework. Originality/value – The research discusses and analyzes the multidimensional facets of AI, as well as the risks that should be assessed and mitigated in organizations, with a focus on country-specific guidelines. © 2026 Emerald Publishing Limited}, keywords = {Artificial intelligence; Ethical considerations; Financial sector; India; UAE}, } @article{vanlierfa2026273, title = {Debate: Beyond the transaction—Procurement as a tool for cross-sector AI governance}, author = {van Lier F.-A.}, year = {2026}, journal = {Public Money and Management}, volume = {46}, number = {1}, pages = {8-9}, doi = {10.1080/09540962.2025.2532174}, abstract = {[No abstract available]}, } @article{shereefdeenhmackaym2026274, title = {“It's like dynamite—It can do a lot of good. It could do a lot of harm”: A qualitative study on the uses, benefits, and risks of genAI in public health}, author = {Shereefdeen H. and MacKay M.}, year = {2026}, journal = {Digital Health}, volume = {12}, doi = {10.1177/20552076261416713}, abstract = {Objective: The adoption of genAI is rapidly evolving all sectors. In public health, emerging genAI technologies have shown promise in facilitating tailored communication, public health surveillance, and administration and decision making. However, adoption of genAI gives rise to several concerns, including the perpetuation of systemic inequities and erosion of public trust. The lack of clear guidelines and directives presents challenges for the responsible integration of genAI. Therefore, this study aims to: (1) explore the understanding of the application of genAI by public health professionals; (2) identify barriers and enablers to responsible genAI use; and (3) explore the perceived governance needs and opportunities for community engagement to guide the responsible and trustworthy implementation. Methods: A semistructured interview guide was iteratively developed, and Canadian public health professionals with experience in genAI technology were recruited via purposive and snowball sampling. All interviews were conducted between 5 June and 21 July 2025. Results: Data from 13 interviews were analyzed using reflexive thematic analysis, from which 7 unique themes emerged: (1) uses of genAI and shift in priorities, (2) emerging skills demands, (3) shift in public health values, data use, and equity, (4) governance imperative, (5) organizational-level guidance, (6) importance of fostering trust, and (7) inclusion of community as co-creators. Conclusion: These themes offer insight into the complexities and challenges of responsible genAI adoption, underscoring the need for governance and organizational frameworks that support equitable, accountable, and transparent implementation within public health. These themes offer guidance to facilitate the responsible integration of genAI in public health and highlight the national organizational governance considerations. © The Author(s) 2026. This article is distributed under the terms of the Creative Commons Attribution-NonCommercial 4.0 License (https://creativecommons.org/licenses/by-nc/4.0/) which permits non-commercial use, reproduction and distribution of the work without further permission provided the original work is attributed as specified on the SAGE and Open Access page (https://us.sagepub.com/en-us/nam/open-access-at-sage).}, keywords = {AI governance; Canada; Generative AI; public health; reflexive thematic analysis}, } @article{bernasconilkapaklikayazgrossmannr2026275, title = {Between promise and practice: Exploring AI’s role in research participant communication}, author = {Bernasconi L. and Kapaklikaya Z. and Grossmann R.}, year = {2026}, journal = {Research Ethics}, doi = {10.1177/17470161261419861}, abstract = {This study examines the use of artificial intelligence (AI) in communication with research participants during the recruitment process, exploring stakeholder perspectives and evaluating ChatGPT’s ability to generate participant informational texts. A mixed-methods design was applied, combining an online survey with semi-structured interviews. The survey, conducted among clinical research professionals and laypersons, investigated perceived suitability, trustworthiness, ethical concerns, and requirements for responsible implementation of AI across use cases such as generating informational texts, visual aids, translations, and chatbot-based interactions. The interviews assessed how experts from diverse disciplines responded to AI-generated versus human-written texts, with particular attention to clarity, empathy, and emotional impact. Survey findings indicated that translation and writing assistance were viewed as the most appropriate and least controversial applications of AI, whereas answering participant questions or assessing eligibility were regarded with caution. AI appears to be rather underused, partly due to regulatory uncertainty and limited trust. Trust in AI was generally lower than trust in humans, especially among professionals. Key ethical concerns included unreliability, data-protection risks, and the potential for manipulation. In the empirical comparison, AI-generated texts were frequently described as clearer, more concise, and more empathetic than human-written materials, yet concerns persisted regarding missing content and oversimplification. Interviewees emphasized the need for strong human oversight, adequate researcher training, and continued involvement of patients in developing participant-facing materials. Maintaining a human role was seen as critical not only for accuracy and accountability, but also to prevent the dehumanization of research communication and preserve communication skills within research teams. Overall, the study highlights both the promise and the ethical complexities of integrating AI into participant communication during recruitment and offers a foundation for future research and policy development. © The Author(s) 2026. This article is distributed under the terms of the Creative Commons Attribution-NonCommercial 4.0 License (https://creativecommons.org/licenses/by-nc/4.0/) which permits non-commercial use, reproduction and distribution of the work without further permission provided the original work is attributed as specified on the SAGE and Open Access page (https://us.sagepub.com/en-us/nam/open-access-at-sage).}, keywords = {AI; AI ethics; clinical research; informed consent; large language models; patient recruitment}, } @article{edmondjusshers2026276, title = {From ‘Hello World’ to ‘How are We Doing?’: fostering collective sensemaking with and about the unidirectional conversationality of code}, author = {Edmond J. and Ussher S.}, year = {2026}, journal = {AI and Society}, doi = {10.1007/s00146-026-02991-1}, abstract = {The manner in which the software platforms and tools that surround us are built and marketed is designed to lead us not into conversations, but silences. It has been 20 years since Cass Sunstein first warned us of the manner in which new media lead us to choose as consumers, rather than citizens, and yet we still find that the dominant paradigm for technology adoption primes us to consider our decision-making around ICT tools as independent and neutral, needing only to meet a measure of value for money. Because of the all-pervasive nature of code and its compiled products, however, this particular socio-technical imaginary brings with it an exceptional power to alienate us not just from each other, but, in a post-Marxian twist, from the very tools, we might use to manage this deterioration in interpersonal exchange. This is, in particular, the case with artificial intelligence, due to the opacity and complexity of the methods deployed, and the dark ethical underbelly of its effects on power, labour, and rights. This article will look at this and other imaginaries that hold us back from forging a cultural relationship to AI and other forms of coded culture that is not just human-centred, but community aware, embodying the oft-cited values of human-centredness and trustworthiness. To do this, it will propose strategies by which to recentre code as a site for the discursive activity of collective sensemaking. Starting from the roots of the human-readable in code, the argument coalesces around the details of two experimental cases that have sought to foster human conversations that include code as an active participant. © The Author(s) 2026.}, keywords = {AI ethics; Code as conversation; Collective sensemaking; STEAM approaches; Artificial intelligence; Ethical technology; AI ethic; Code as conversation; Collective sensemaking; New media; Power; Sense making; Software platforms; Software-tools; STEAM approach; Technology adoption; Codes (symbols)}, } @article{cohenmkhavkinmmovsowitzdavidowdtoche2026277, title = {ChatGPT in the public eye: Ethical principles and generative concerns in social media discussions}, author = {Cohen M. and Khavkin M. and Movsowitz Davidow D. and Toch E.}, year = {2026}, journal = {New Media and Society}, volume = {28}, number = {1}, pages = {5-31}, doi = {10.1177/14614448241279034}, abstract = {With ChatGPT’s rapid adoption, concerns regarding generative artificial intelligence (AI) have shifted from theoretical to practical. Drawing upon the “algorithmic imaginary” framework from critical algorithm studies and the anthropological concept of “ordinary ethics,” we analyzed Twitter discourse during ChatGPT’s initial deployment, examining 368,359 tweets. Our analysis identified five topics reflecting functional and critical aspects of ChatGPT. We specifically point to two topics with a critical perspective: “Ethics” and “Concerns.” The first aligns with scholarly discussions in AI ethics on fairness and transparency, while the second focuses on ChatGPT’s generative capabilities. This highlights an emerging trend: While the academic discussion on AI ethics has gained popularity, especially in scrutinizing ChatGPT, the conversation is now expanding to more nuanced ethical deliberations. We analyzed the posts’ engagement and sentiment over time, demonstrating the AI ethics community’s influence in addressing the potential and harms of generative AI systems. © The Author(s) 2024. This article is distributed under the terms of the Creative Commons Attribution-NonCommercial 4.0 License (https://creativecommons.org/licenses/by-nc/4.0/) which permits non-commercial use, reproduction and distribution of the work without further permission provided the original work is attributed as specified on the SAGE and Open Access page (https://us.sagepub.com/en-us/nam/open-access-at-sage).}, keywords = {AI ethics; algorithm; algorithmic imaginary; ChatGPT; discourse; early adoption; generative AI; ordinary ethics; Twitter; user agency}, } @article{leejpangck2026278, title = {A YIN‑YANG APPROACH TO BALANCED AI REGULATION AND LESSONS FOR SINGAPORE}, author = {Lee J. and Pang C.K.}, year = {2026}, journal = {Singapore Academy of Law Journal}, volume = {38}, pages = {190-246}, abstract = {Lao Tzu’s quote – “[a]ll things carry yin yet embrace yang. They blend their life breaths in order to produce harmony”1 – carries lessons on striking the right balance for AI regulation in Singapore. For Singapore, which is a “fast follower” in the global AI race, the key is to learn lessons quickly from international precedents, while aligning with and influencing global developments to ensure interoperability. This article aims to demonstrate that Singapore should not seek to find a single “correct” point of balance. Instead, it should advance aspects key to building the groundwork for international AI governance – and in so doing, serve as a forward-looking model for how small states can ride the AI wave. © 2026 Contributor(s) and Singapore Academy of Law.}, } @article{alsudaism2026279, title = {EVOLVING CORPORATE FIDUCIARY DUTIES AND LIABILITIES IN THE ERA OF ARTIFICIAL INTELLIGENCE-DRIVEN SUSTAINABILITY}, author = {Alsudais M.}, year = {2026}, journal = {Corporate Governance and Sustainability Review}, volume = {10}, number = {3}, pages = {110-140}, doi = {10.22495/cgsrv10i3p9}, abstract = {This study examines how the integration of artificial intelligence (AI) into corporate operations fundamentally reshapes the core doctrines of corporate law and governance. Our analysis establishes that AI represents a structural transformation that necessitates a complete re-evaluation of traditional fiduciary duties, liability frameworks, and supervisory approaches. The research highlights AI’s dual impact on sustainability: its immense potential to advance environmental, social, and governance (ESG) objectives (through enhanced measurement and transparency) is directly counterbalanced by significant risks, including algorithmic discrimination, environmental externalities from high computing intensity, and opaque decision-making. The core finding is that existing corporate governance frameworks inadequately address this dual and contradictory impact. Consequently, the study concludes that a fundamental doctrinal evolution is essential. Specifically, the analysis identifies the imperative to extend fiduciary obligations to encompass technological competence, ethical oversight, and data stewardship—a necessary concept framed here as the “duty of technological prudence”. Furthermore, the research demonstrates how AI exposes firms to amplified liabilities, ranging from privacy claims and cybersecurity failures to complex product liability for autonomous systems. To address the fragmented global regulatory landscape, the paper advances “sustainability-by-design” as a strategic imperative, mandating that ethical and environmental considerations be embedded throughout the entire AI lifecycle. The paper concludes with targeted policy recommendations for legislators and corporate boards to align AI innovation with social progress and human dignity. © 2026 The Author.}, keywords = {AI Governance; Corporate Liability; Data Stewardship; ESG; Fiduciary Duties; Sustainability-by-Design; Technological Prudence}, } @article{tanzengjliz2026280, title = {Governance of discriminatory content in conversational AIs: a cross-platform and cross-cultural analysis}, author = {Ta N. and Zeng J. and Li Z.}, year = {2026}, journal = {Information Communication and Society}, volume = {29}, number = {6}, pages = {1855-1872}, doi = {10.1080/1369118X.2025.2537803}, abstract = {As the widespread adoption of conversational artificial intelligence (AI) systems has raised concerns about social bias, especially towards vulnerable groups, this study explores how these systems respond to and regulate discriminatory content. Through a cross-lingual, cross-platform comparative analysis, this study systematically examines six leading conversational AI systems: ChatGPT, Gemini, Llama, Ernie Bot, ChatGLM, and Tongyi. Using a mixed-method approach, the study reveals that refusal sensitivity and answering strategies vary significantly across systems, languages, and topics. This paper provides new insights into the moderation strategies of conversational AI systems and introduces a framework for auditing AI content moderation concerning social discrimination. © 2025 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group.}, keywords = {AI bias; AI governance; computational methods; Conversational AI; cross-cultural analysis}, } @article{dumitriup2026281, title = {The emergence of international norms on AI ethics}, author = {Dumitriu P.}, year = {2026}, journal = {World Journal of Science, Technology and Sustainable Development}, volume = {21}, number = {1/2/3}, pages = {27-48}, doi = {10.47556/J.WJSTSD.21.1-2-3.2026.3}, abstract = {PURPOSE: This study aims to analyse the perspectives of the United Nations (UN) system, European Union (EU) and other international organisations on the ethical aspects of the use of Artificial Intelligence (AI) tools. DESIGN/METHODOLOGY/APPROACH: The paper will explore the attempts to define the main ethics challenges that may accompany the use of AI as reflected in the documents adopted by international organisations, in particular by the EU and the UN. FINDINGS: As the UN does not have its own capacity to develop AI tools, it is playing a considerable role in defining principles, limits, and possible rules on the use of AI and stimulating international co-operation. However, the regulations adopted by the EU on AI may influence international co-operation at the global level. ORIGINALITY: The paper identifies the most significant actions taken at international level, and analyses their similarities, differences, and neglected areas in dealing with the preservation of a strong ethical dimension and the use of AI in service of the public good. IMPLICATIONS: The paper will help to develop a better knowledge and understanding of the most meaningful conceptual advances in defining and disseminating the principles of ethics among the many stakeholders involved in developing, promoting, and using AI instruments. © 2026 by all the authors of the article above. The article is published as an open access article by WASD under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).}, keywords = {AI; Council of Europe Framework Convention; Enforcement; Ethics; EU AI Act; UNESCO; Unethical Practices; WSIS}, } @article{akiniwilczkowskalakinm2026282, title = {Ethical and Data Privacy Risks in AI-Driven Accounting: A Mixed-Methods Study}, author = {Akin I. and Wilczkowska L. and Akin M.}, year = {2026}, journal = {Human Behavior and Emerging Technologies}, volume = {2026}, number = {1}, pages = {6048044}, doi = {10.1155/hbe2/6048044}, abstract = {Artificial intelligence (AI) is increasingly integrated into accounting, automating tasks such as data processing, risk assessment and financial reporting. While AI offers efficiency gains, it raises ethical and data privacy risks, challenges professional accountability and tests the effectiveness of governance frameworks. This study investigates how these risks are perceived, managed and governed in AI-driven accounting environments. A mixed-methods approach was adopted, combining semistructured interviews and surveys with 24 accounting professionals in the Bristol and Bath area. The findings indicate that AI system opacity, algorithmic bias and sensitive data handling are key ethical concerns, while accountability depends on both organisational mechanisms and individual experience with AI. Governance and regulatory alignment support trust but are insufficient without ethical embedding and professional oversight. Based on these insights, the study recommends enhancing AI transparency, implementing clear accountability frameworks and embedding ethics into organisational culture. Copyright © 2026 Isik Akin et al. Human Behavior and Emerging Technologies published by John Wiley & Sons Ltd.}, keywords = {AI in accounting; data privacy; decision-making; ethics; risk mitigation}, } @article{jaworskik2026283, title = {Ethics of Artificial Intelligence: Old Moral Questions and Transformative Challenges}, author = {Jaworski K.}, year = {2026}, journal = {Theology and Science}, volume = {24}, number = {2}, pages = {382-399}, doi = {10.1080/14746700.2026.2637221}, abstract = {Artificial intelligence (AI) transcends human capacities in computational prowess and decision-making, enriching science, culture, and industry. Yet it provokes urgent ethical questions: from algorithmic bias and privacy breaches to intellectual property disputes and opaque decision-making, as well as automation’s disruption of labor markets. Its infiltration into art, literature, and cultural production challenges assumptions about creativity and human agency. This article scrutinizes seven key domains—cybersecurity and disinformation, employment, bias, transparency, intellectual property, privacy, and aesthetics—to assess whether AI’s ethical quandaries are genuinely novel or modern iterations of entrenched moral concerns. Ultimately, it advocates re-examining these issues amidst the Fourth Industrial Revolution. © 2026 Graduate Theological Union (CTNS Program).}, keywords = {AI; artificial intelligence; black box; cybersecurity; discrimination; employment; ethics}, } @article{setiantoypoctaviantoawjallin2026284, title = {Imaginaries of precarity: negotiating generative AI in Indonesia’s cultural and creative labor}, author = {Setianto Y.P. and Octavianto A.W. and Jalli N.}, year = {2026}, journal = {AI and Society}, doi = {10.1007/s00146-026-03121-7}, abstract = {This paper examines how cultural and creative workers in Indonesia negotiate the adoption of Generative Artificial Intelligence (GenAI) under conditions of labor precarity. Drawing on sociotechnical imaginaries and scholarship on precarious creative labor, the study explores how workers interpret, appropriate, and constrain GenAI in their everyday practices. The research adopts a Participatory Action Research (PAR) approach in collaboration with a creative workers’ union, Serikat Sindikasi, combining focus group discussions, in-depth interviews, and a feedback workshop involving 40 participants across multiple subsectors. The findings identify three sociotechnical imaginaries: GenAI as a bounded assistant, through which workers seek to maintain authorship and professional identity; GenAI as an engine of acceleration, intensifying productivity demands and reshaping temporal expectations; and GenAI as a contested system of governance, raising concerns about fairness, ownership, and regulation. Together, these imaginaries show that GenAI adoption is deeply shaped by existing labor insecurities. The paper argues that workers’ selective engagement with GenAI reflects efforts to navigate precarity, underscoring the need for context-sensitive, labor-centered approaches to AI governance in the Global South. © The Author(s), under exclusive licence to Springer-Verlag London Ltd., part of Springer Nature 2026.}, keywords = {Artificial intelligence; Cultural and creative industry; Indonesia; Precarious labor; Sociotechnical imaginaries; Employment; Industrial relations; Personnel; Condition; Creatives; Cultural and creative industries; Imaginaries; Indonesia; Participatory action research approaches; Precarious labor; Sociotechnical; Sociotechnical imaginarie; Workers'; Artificial intelligence}, } @article{liolioueseitanidimmstadtlerl2026285, title = {Governance under Algorithmic Opacity: How Financial Firms Construct Accountability and Control around AI in Risk Disclosures}, author = {Lioliou E. and Seitanidi M.M. and Stadtler L.}, year = {2026}, journal = {Information Systems Frontiers}, doi = {10.1007/s10796-026-10762-y}, abstract = {As artificial intelligence (AI) systems reshape financial decision-making, firms face increasing pressure to assure regulators and investors that they remain in control of new technologies that are inherently opaque and difficult to predict. This study examines how financial institutions construct accountability around AI through mandatory risk disclosures. Using a three-year panel of 10-K filings from 73 publicly listed S&P financial firms, we combine keyword-based extraction with supervised machine learning to classify disclosure language along a continuum of AI accountability framing. This continuum ranges from locating accountability within the firm’s own governance structures, to situating it within external institutional rules and to dispersing it through technological unpredictability and partial autonomy. Our study advances research on responsible AI governance by showing that AI accountability is not only engineered through formal controls but also constructed discursively, as firms use accountability framings to locate and shape the boundaries of control under conditions of algorithmic opacity. Complementing accounts of AI unpredictability as a sociotechnical condition, our findings show that firms pair claims of control with disclosures of “known unknowns”, presenting AI risks as governance-relevant yet not fully governable. In this manner, firms actively construct what accountability can mean when full explainability is unattainable. We also document how the prevalence of these framings shifts over time, consistent with intensifying regulatory scrutiny and the rise of generative AI. © Crown 2026.}, keywords = {Accountability framings; AI accountability; AI governance; Algorithmic opacity; Responsible AI; Artificial intelligence; Decision making; Investments; Accountability framing; Algorithmic opacity; Algorithmics; Artificial intelligence accountability; Artificial intelligence governance; Artificial intelligence systems; Condition; Decisions makings; Financial decisions; Responsible artificial intelligence; Opacity}, } @article{wangqqiylir2026286, title = {Artificial Intelligence and Corporate Sustainability: Shaping the Future of ESG in the Age of Industry 5.0}, author = {Wang Q. and Qi Y. and Li R.}, year = {2026}, journal = {Sustainable Development}, volume = {34}, number = {1}, pages = {1-26}, doi = {10.1002/sd.70200}, abstract = {The growing concern about the disruptive potential of new technologies compels us to reconsider the impact of AI on sustainability. This study examines the development of AI within Chinese listed companies from two key perspectives: technology-driven innovation and institutional responses, and analyzes their effects on corporate sustainability. The findings reveal that (i) AI significantly enhances corporate sustainability, particularly within the tertiary industry; (ii) both productivity and information transparency serve as positive mediators in this relationship; (iii) employee education plays a crucial role, with a higher proportion of employees holding bachelor's and master's degrees benefiting firms more significantly; (iv) at the institutional level, the degree of factor market development exhibits a notable heterogeneous effect, with cities having lower levels of factor market development showing a stronger contribution of AI to sustainability—likely due to the diminishing marginal effect of technology; and (v) proactive AI policies are instrumental in improving economic sustainability, underscoring the positive role of well-targeted government intervention. This study offers both technical and policy insights for governments seeking to foster AI-driven sustainability while ensuring that AI itself contributes to broader sustainable development goals. © 2025 ERP Environment and John Wiley & Sons Ltd.}, keywords = {artificial intelligence (AI); corporate sustainability; employee education; ESG (environmental, social, and governance); policy intervention; technology-driven innovation; China; advanced technology; artificial intelligence; corporate strategy; future prospect; industrial technology; innovation; Sustainable Development Goal}, } @article{kailaakkotsiosaholzapfela2026287, title = {Terms of Fairness in Synthetic Singing}, author = {Kaila A.-K. and Kotsios A. and Holzapfel A.}, year = {2026}, journal = {IEEE Transactions on Technology and Society}, doi = {10.1109/TTS.2026.3687375}, abstract = {The fast proliferation of AI-driven singing voice synthesis (SVS) facilitates wide-scale misappropriation of third-party vocal expressions, while sidelining the interests of practicing vocal artists. Since EU law provides limited protection for the human singing voice, the emerging SVS industry practices and norms are mainly self-regulated through the Terms of Service (ToS). In this study, we survey the ToS of ten commercial SVS services to examine how they respond to the intensifying calls for Responsible AI and articulate their understanding of a fair market. This is the first paper that examines the contractual dispositions of the service providers of generative AI models and their ethical implications in generative music applications for singing. Using the lenses of procedural and substantive fairness, we chart how service providers allocate rights and obligations among the stakeholders and outline the forms of agency the voice artist can be assigned despite the legislative void. Our work contributes to the socio-technical study of fairness in the context of the developing SVS market. It also informs generative AI policies and further advocacy efforts to cultivate responsible innovation and defend the integrity of vocal artists against synthetic exploitations. © 2020 IEEE.}, keywords = {AI ethics; AI music; Fairness; generative AI; responsible AI; singing voice synthesis; terms of service; vocal artist; Ethical technology; Laws and legislation; Service industry; AI ethic; AI music; Fairness; Generative AI; Responsible AI; Service provider; Singing-voice synthesis; Terms of services; Third parties; Vocal artist; Commerce}, } @article{vallrsauptokanbachdkkrauss2026288, title = {The role of formal and informal mechanisms in AI-driven new product development}, author = {Vallé R. and Saup T.-O. and Kanbach D.K. and Kraus S.}, year = {2026}, journal = {Management Decision}, pages = {1-19}, doi = {10.1108/MD-10-2025-3324}, abstract = {Purpose – This paper explores how organizations can effectively govern artificial intelligence (AI) in new product development (NPD), focusing on the interplay between formal and informal mechanisms. While AI enhances creativity and efficiency in NPD, its emergent and informal use creates risks related to innovation outcomes. Our study investigates how organizations manage these tensions over time and how governance evolves as AI becomes embedded in NPD. Design/methodology/approach – We conducted a qualitative multiple-case study. Data were collected from 38 semi-structured interviews and supplementary material across 10 companies representing different sizes and products. The data were analyzed through iterative qualitative coding, progressing from first-order concepts to second-order themes and aggregated dimensions (Gioia methodology). Findings – The study develops a dynamic framework of AI governance in NPD comprising four dimensions: AI adoption in NPD, formal mechanisms, informal mechanisms, and dynamic governance for AI in NPD. Findings show that formal controls are typically introduced to mitigate uncertainty but often coexist with informal mechanisms. Effective AI governance emerges from the continuous reconfiguration of enabling and constraining mechanisms rather than from static control. Originality/value – We conceptualize AI governance in NPD as an evolving configuration of formal and informal mechanisms. By integrating organizational control theory and the Levers of Control framework, it advances a dynamic perspective on AI governance and explains how organizations can manage the tension between innovation autonomy and control in AI-driven NPD contexts. © Emerald Publishing Limited}, keywords = {AI governance; Artificial intelligence; Formal and informal mechanisms; Innovation management; New product development}, } @article{hasibmekol2026289, title = {Between Duty and Discipline: Kantian Ideals and Techno-Governmental Realities in the European Union Artificial Intelligence Act and the United Nations Artificial Intelligence Framework}, author = {Hasib M. and Eko L.}, year = {2026}, journal = {International Journal of Communication}, volume = {20}, pages = {1190-1213}, doi = {10.65476/zxewbx77}, abstract = {This study comparatively analyzes two major global AI governance frameworks—the European Union's AI Act and the United Nations' Governing AI for Humanity report—through the dual lenses of Kantian ethics and techno-governmentality. While previous research has focused on the ethical content of these policies, this study investigates the fundamental tension between their normative rhetoric and operational mechanisms. The findings reveal a critical paradox: While both documents articulate strong Kantian commitments to human dignity, autonomy, and rights, their implementation strategies rely heavily on techno-governmental logics—managing populations through surveillance, classification, and statistical risk assessment. The analysis highlights deep structural constraints, including the EU's reliance on internal market competence, which subordinates rights to communitarian market standardization, and the "responsibility gap" in autonomous systems, rendering strict Kantian duty ethics structurally unfulfillable. Ultimately, the study argues that without robust participatory mechanisms, ethical AI governance risks becoming a tool shaped by power dynamics and bureaucratic control rather than moral protection. This research concludes that restoring human autonomy requires moving beyond technical compliance to democratically legitimized, inclusive decision making. Copyright © 2026 (Mir Hasib and Lyombe Eko). Licensed under the Creative Commons Attribution Non-commercial No Derivatives (by-nc-nd). Available at https://ijoc.org.}, keywords = {AI policy; artificial intelligence; EU Kantian AI ethics; European Union AI regulation; techno-governmentality in AI; UN risk-based AI regulation; United Nations AI report}, } @article{oloyeder2026290, title = {Collaborative regulation: leveraging existing authorities for effective AI governance in Africa}, author = {Oloyede R.}, year = {2026}, journal = {International Review of Law, Computers and Technology}, volume = {40}, number = {2}, pages = {242-261}, doi = {10.1080/13600869.2025.2506916}, abstract = {Artificial intelligence (AI) is gradually evolving in Africa, presenting unique regulatory challenges, including its potential ability and potential to disrupt industries, exacerbate inequalities, and impact human rights. While calls for new AI-specific laws and the creation of new agencies are widespread, financial constraints, limited institutional capacity, and the lack of political will often hinder such efforts. This paper proposes a pragmatic and timely solution: leveraging existing regulatory bodies to govern AI within their current mandates. These authorities, already established in many African countries, possess the institutional capacity to address AI’s sector-specific risks, bypassing the need for costly new entities. This paper analyses the activities of these authorities in regulating AI across Africa, highlighting their successes, challenges, and opportunities for further action. The paper advocates for a collaborative approach using a digital clearinghouse model where regulators share insights and harmonise strategies. While acknowledging that new, AI-specific regulations may eventually be necessary, this approach offers an immediate and effective path towards AI governance in Africa, ensuring that AI’s benefits are harnessed responsibly and inclusively for sustainable development across the continent. © 2025 Informa UK Limited, trading as Taylor & Francis Group.}, keywords = {Africa; Artificial intelligence; clearing house model}, } @article{kaclovm2026291, title = {Adopting AI in transitional contexts: Professionalization and ethical dilemmas in Central European public relations}, author = {Kaclová M.}, year = {2026}, journal = {Communications}, doi = {10.1515/commun-2025-0055}, abstract = {This study explores how public relations professionals in three post-authoritarian countries - the Czech Republic, Slovakia, and Poland - navigate the adoption of artificial intelligence (AI) tools in their daily work. It studies perceived opportunities and barriers as well as normative understandings of AI use among PR professionals. Drawing on the Transitional Public Relations Model and Professionalization Theory, the study examines how institutional legacies and democratic fragilities shape both ethical norms and professional practices. Based on 16 semi-structured interviews and a hybrid thematic analysis, the findings reveal a paradox: AI is embraced for its utility in routine tasks, yet its use unfolds in underregulated environments with limited training and persistent distrust in institutions. Formal ethical frameworks for AI use are largely absent, leaving practitioners to rely on individual judgement. This research advances the understanding of technology adoption, especially in transitional democracies, and highlights the impact of systemic factors on responsible AI integration in communication practice. It contributes a region-specific perspective to global debates on AI ethics, digital transformation, and the evolving role of communication professionals. © 2026 Walter de Gruyter GmbH, Berlin/Boston.}, keywords = {AI; Communication; ethics; professionalization; public relations; Transitional PR Theory}, } @article{cleofasjv2026292, title = {Nursing Professional Organisations as Human Rights Intermediaries: Towards an Integrated Framework of Stakeholdership for Healthcare AI Governance}, author = {Cleofas J.V.}, year = {2026}, journal = {Journal of Clinical Nursing}, doi = {10.1111/jocn.70256}, abstract = {Aim: To propose a normative framework that guides nursing professional organisations to act as human rights intermediaries in the governance of artificial intelligence in healthcare. Design: Discursive paper. Results: The paper presents a triaxial framework that conceptualises the role of nursing professional organisations in artificial intelligence governance. The framework consists of a domain axis, which identifies key areas of engagement; a modality axis, which aligns actions with the specific functions of these organisations; and a human rights axis, which defines their role towards rights claimants and duty bearers. Conclusion: The proposed framework provides a practical tool for nursing professional organisations to strategically plan and implement initiatives to influence the advancement and regulation of artificial intelligence. Its application can help ensure that healthcare innovation is equitable and rights-based. Implications for the Profession: This paper provides a blueprint for nursing leaders and policymakers to engage proactively with the ethical dimensions of artificial intelligence. It emphasises the salient roles of nursing professional organisations in advocating for the human right to health in a technologically driven healthcare landscape. Impact: This paper addresses the gap in how the nursing profession can systematically engage with artificial intelligence governance. The main finding is a novel framework that provides a structured way for nursing professional organisations to act as human rights intermediaries. This research will have a significant impact on nursing leadership, patient advocacy groups, and policymakers involved in healthcare technology and ethics. Patient or Public Contribution: Initial parts of this paper were presented to allied health practitioners via a webinar, providing early feedback and dialogue that informed its development. © 2026 John Wiley & Sons Ltd.}, keywords = {artificial intelligence; delivery of health care; health care governance; human rights; nursing; nursing professional organization; advocacy group; article; artificial intelligence; diagnosis; feedback system; health care; health care delivery; health practitioner; human; human rights; leadership; medical society; nursing; nursing as a profession; occupation; patient advocacy; right to health; webinar}, } @article{hecppanyrchukc2026293, title = {AIDLC–FMEA: Explainability-Integrated Lifecycle Risk Assessment for AI Governance}, author = {He C.-P. and Pan Y.-R. and Chu K.-C.}, year = {2026}, journal = {IEEE Access}, doi = {10.1109/ACCESS.2026.3701925}, abstract = {The growing deployment of AI-enabled systems in safety- and decision-critical contexts has exposed limitations in existing governance and risk assessment approaches, particularly regarding lifecycle alignment, interpretability, and decision support. This paper presents AIDLC-FMEA, a generalizable and explainability-aware governance risk analysis framework that extends Failure Mode and Effects Analysis to support structured examination of AI governance across the development lifecycle. The framework formulates governance risk as a phase-coupled aggregation problem and incorporates explainability as a sensitivity-based risk amplification mechanism within a lifecycle mapping structure. An Extended Governance Risk Priority Number is defined to characterize phase-level governance-related risk, which is subsequently aggregated into bounded system-level compliance indices to support comparative analysis and interpretive benchmarking in relation to selected governance frameworks. The framework is evaluated through formal analysis of key model properties, simulation-based sensitivity analysis, and illustrative reconstruction of post-2020 AI incident cases. These analyses illustrate how lifecycle-aware risk aggregation and explainability-related sensitivity contribute to differentiated governance risk profiles across heterogeneous AI systems, in ways that can be interpreted in relation to selected governance concepts reflected in ISO/IEC 42001. Overall, AIDLC-FMEA is designed to support governance risk diagnosis and comparative analysis, rather than prediction or certification, and provides an interpretable and computationally tractable analytical tool for enterprise-level AI governance and managerial risk oversight. © 2013 IEEE.}, keywords = {AI development lifecycle; AI governance; comparative governance analysis; engineering management; explainability; Failure Mode and Effects Analysis; governance risk prioritization; lifecycle-based risk analysis; Artificial life; Decision support systems; Failure (mechanical); Formal concept analysis; Life cycle; Life cycle assessment; Risk analysis; Risk assessment; Risk management; Safety engineering; AI development lifecycle; AI governance; Comparative governance analyze; Engineering management; Explainability; Failure mode and effects analysis; Governance risk prioritization; Governance risks; Lifecycle-based risk analyze; Risk analyze; Risk prioritization; Sensitivity analysis}, } @article{frizzomcolleonieromentis2026294, title = {Governing the Issue: Strategic Delayed Moralization of AI in Law Enforcement}, author = {Frizzo M. and Colleoni E. and Romenti S.}, year = {2026}, journal = {Business and Society}, doi = {10.1177/00076503261447590}, abstract = {Artificial intelligence (AI) technologies are typically first adopted as embodiments of rationality and neutrality, and only later do their moral implications receive public scrutiny. This temporal gap between adoption and moral recognition has been widely observed, yet how organizations actively sustain it remains undertheorized. This study theorizes strategic delayed moralization as the process through which organizations actively shape the timing and trajectory of AI transition from technical artifacts to morally recognized objects. We focus on ShotSpotter, an AI-powered gunshot detection system used in law enforcement, analyzing a decade of organizational communications and public discourse. Combining issue management and strategic ambiguity theory, we develop a process model showing how organizations deploy evolving forms of strategic ambiguity to delay moral recognition, and how this process is ultimately constrained by societal moralization that progressively narrows the space for ambiguity. This research contributes to research on AI in society by showing how business and society interact in shaping the moral trajectory of AI. © The Author(s) 2026}, keywords = {AI governance; artificial intelligence; issue management; moralization of technology; ShotSpotter; strategic ambiguity}, } @article{bentalebd2026295, title = {She codes, she leads: the tipping point for ethical digital transformation in banking}, author = {Bentaleb D.}, year = {2026}, journal = {Journal of Information, Communication and Ethics in Society}, pages = {1-22}, doi = {10.1108/JICES-09-2025-0246}, abstract = {Purpose – The purpose of this study is to examine the nonlinear threshold effect of board and executive gender diversity on ethical digital transformation (EDT) in European banks. While gender diversity is increasingly promoted as a governance mechanism for responsible innovation, its impact on the ethical quality of digital transformation remains ambiguous, particularly below critical levels of representation. This paper identifies the precise tipping points at which female leadership becomes a catalyst for fair, inclusive and accountable digital innovation. Design/methodology/approach – Using a panel data set of 224 European banks over the period 2011–2022, the authors construct a novel composite index of EDT based on three dimensions: AI ethics and algorithmic fairness, digital inclusion and financial accessibility and user-centric design and accessibility. The Panel Smooth Transition Regression (PSTR) model is used to estimate endogenous thresholds and capture the nonlinear, regime-dependent relationship between gender diversity and EDT, while controlling for board characteristics (size, independence, digital and sustainability expertise) and bank-specific factors (size, capital, profitability and risk). Findings – The results of this study reveal a significant critical mass effect. Board gender diversity enhances EDT only when women represent more than 28.70% of board members; below this threshold, the effect is statistically insignificant. Similarly, executive gender diversity significantly improves EDT only beyond a threshold of 52.10%, underscoring the strategic influence of women in operational leadership roles (e.g. Chief Technology Officer and Chief Digital Officer). These findings support critical mass theory and upper echelons theory, demonstrating that women in executive positions directly shape ethical technology governance. This study also shows that the effect is amplified by board digital expertise and independent oversight. Research limitations/implications – The sample is limited to European banks, and endogeneity, particularly reverse causality, cannot be fully ruled out. Future research could use instrumental variables or qualitative methods to explore causal mechanisms. Practical implications – The identified thresholds provide empirically grounded benchmarks for policymakers (e.g. ECB and EBA) and financial institutions seeking to embed gender-inclusive governance into digital accountability frameworks such as the AI Act and CSRD. Banks should move beyond symbolic representation and target substantive female leadership in both governance and executive functions to ensure technology-driven accountability. Originality/value – To the best of the authors’ knowledge, this study is among the first to apply the PSTR model to examine the nonlinear dynamics of gender diversity in the context of EDT. It introduces a robust, multidimensional EDT index and provides endogenous estimates of critical mass thresholds, bridging gender diversity theory, responsible innovation and sustainable digital governance in the financial sector. © 2026 Emerald Publishing Limited}, keywords = {Critical mass; Digital inclusion; Ethical digital transformation; European banks; Gender diversity; PSTR model; Responsible AI; Sustainable governance}, } @article{alsaighrmehmoodrkatibialmuzainiaaalbouqssalshmranys2026296, title = {Evidence-Driven AI Governance for Healthcare: A PEARL-PATHWAY Analysis of Madinah}, author = {Alsaigh R. and Mehmood R. and Katib I. and Almuzaini A.A. and Albouq S.S. and Alshmrany S.}, year = {2026}, journal = {International Journal of Advanced Computer Science and Applications}, volume = {17}, number = {4}, pages = {198-214}, doi = {10.14569/IJACSA.2026.0170421}, abstract = {The rapid integration of artificial intelligence (AI) into healthcare systems has intensified the need for governance frameworks that ensure safety, accountability, ethical, and sustainable deployment. However, existing AI governance approaches are primarily articulated through high-level ethical and regulatory principles, with limited operational guidance tailored to specific healthcare contexts. This challenge is particularly evident in dynamic settings such as Al Madinah, Saudi Arabia, where demographic diversity, evolving healthcare needs, and large-scale public health pressures, including the presence of millions of visitors annually during Hajj and Umrah, require adaptive and context-aware governance. This study presents an evidence-driven approach to AI governance analysis that directly links empirical healthcare needs with regulatory frameworks. It integrates the PEARL framework to systematically analyse an initial corpus of 4,277 healthcare publications related to Madinah, refined to 243 articles through inclusion and exclusion criteria, extracting structured representations of healthcare priorities, with the PATHWAY framework to evaluate alignment between these needs and both Saudi Arabian and international AI governance frameworks. This enables a systematic assessment of governance applicability, identification of gaps, and analysis of associated risks. The results reveal that while existing frameworks provide strong foundations in terms of privacy, ethics, and risk-based regulation, they lack operational pathways tailored to domain-specific healthcare requirements and local contexts. Key gaps are identified in areas including epidemiological surveillance, behavioural health, maternal and paediatric care, environmental health integration, and generative AI in public health communication. By bridging empirical evidence with governance analysis, this study advances a structured approach to domain-informed and context-sensitive AI governance. It contributes to the emerging field of computational policy analysis and provides evidence-driven insights for developing adaptive, scalable, and trustworthy AI governance strategies in healthcare systems. © 2026, The Author(s).}, keywords = {AI governance; context-aware governance; domain-specific governance; evidence-driven governance; governance gap analysis; healthcare AI; PATHWAY framework; PEARL framework; policy intelligence; sustainability; Artificial intelligence; Behavioral research; Ethical aspects; Gems; Health care; Public health; Public policy; Risk analysis; Risk assessment; Artificial intelligence governance; Context-Aware; Context-aware governance; Domain specific; Domain-specific governance; Evidence-driven governance; Gap analysis; Governance gap analyze; Healthcare artificial intelligence; PATHWAY framework; PEARL framework; Policy intelligence; Sustainable development}, } @article{ahmedsyaakubanjaveda2026297, title = {The Politics of Artificial Intelligence in Pakistan: Sovereignty, Systemic Fragility and the Battle for Digital Infrastructure}, author = {Ahmed S. and Yaakub A.N. and Javed A.}, year = {2026}, journal = {Perspectives on Global Development and Technology}, doi = {10.1163/15691497-20262007}, abstract = {Artificial intelligence (AI) is becoming integral to statecraft, yet its adoption in politically complex contexts like Pakistan is highly challenging. Against this backdrop, this study investigates the fragmented trajectory of AI integration in Pakistan’s cyberspace, drawing on eighteen experts’ interviews and documentary analysis. It reveals that AI governance is shaped less by coherent strategy and more by bureaucratic competition, foreign technological dependency and patronage politics. Despite a proliferation of policy frameworks and international partnerships, AI readiness remains largely rhetorical, undermined by siloed institutions, fragmented authority, and performative reform. The article identifies four interlocking dynamics: AI as contested political infrastructure, readiness as narrative rather than capacity, institutional turf wars and the outsourcing of core technical competencies. Rather than fostering resilience, AI in weak governance settings risks entrenching opacity and external control. This study contributes to critical debates on digital sovereignty and postcolonial techno-politics, urging a reimagining of who governs AI, how, and to what ends. © Sajjad Ahmed et al., 2026. Published with license by Koninklijke Brill BV. This work is published by Koninklijke Brill BV. Koninklijke Brill BV incorporates the imprints Brill, Brill Nijhoff, Brill Schöningh, Brill Fink, Brill mentis, Brill Wageningen Academic, Vandenhoeck & Ruprecht, Böhlau and V&R unipress. Koninklijke Brill BV reserves the right to protect the publication against unauthorized use and to authorize dissemination by means of offprints, legitimate photocopies, microform editions, reprints, translations, and secondary information sources, such as abstracting and indexing services including databases. Requests for commercial re-use, use of parts of the publication, and/or translations must be addressed to Koninklijke Brill BV.}, keywords = {artificial intelligence; digital sovereignty; state capacity; techno-politics}, } @article{gillespietshawrgraymlsuhj2026298, title = {AI Red-Teaming Is a Sociotechnical Problem}, author = {Gillespie T. and Shaw R. and Gray M.L. and Suh J.}, year = {2026}, journal = {Communications of the ACM}, volume = {69}, number = {2}, pages = {88-95}, doi = {10.1145/3731657}, abstract = {As generative AI technologies find more and more real-world applications, the importance of testing their performance and safety is paramount. “Red-teaming” has quickly become the primary approach to testing AI models—prioritized by AI companies, and enshrined in AI policy and regulation. Members of red teams act as adversaries, probing AI systems to test their safety mechanisms and uncover vulnerabilities. Yet we know far too little about this work or its implications. In this article, we highlight the importance of understanding the values and assumptions behind red-teaming, the labor arrangements involved, and the psychological impacts on red-teamers, drawing insights from lessons learned around the work of content moderation. Red-teaming should be a deeply interdisciplinary concern. To avoid repeating the mistakes of the recent past, we call for a coordinated network of scholars, from the full range of the computational and social sciences, to study the technical, social, critical, and policy dimensions of red-teaming and of the emerging sociotechnical system that is AI. Beyond its utility in detecting AI vulnerabilities and bias, red-teaming raises important issues around values, labor, and harms. © 2025 ACM.}, keywords = {Personnel; AI systems; AI Technologies; Performance; Policy and regulation; Real-world; Red teaming; Safety mechanisms; Sociotechnical; Sociotechnical systems; Social sciences computing}, } @article{tirottaddibellofmontagnania2026299, title = {Clinical governance of artificial intelligence in internal medicine: a literature-informed five-pillar framework for complex care}, author = {Tirotta D. and Di Bello F. and Montagnani A.}, year = {2026}, journal = {Internal and Emergency Medicine}, doi = {10.1007/s11739-026-04408-9}, abstract = {Most artificial intelligence (AI) governance frameworks in healthcare address model development, reporting standards, or regulation in broad terms, but do not adequately translate these principles into the operational realities of internal medicine. Although multimorbidity, frailty, polypharmacy, diagnostic uncertainty, incomplete data, and transitions of care are not exclusive to internal medicine, this specialty is characterized by its frequent and simultaneous convergence within the same clinical decision-making process. We developed a literature-informed conceptual framework through a targeted narrative review and interpretive synthesis of methodological, regulatory, implementation, and clinical literature on AI in healthcare. We examined five source domains: reporting and evaluation standards; risk-of-bias and model-appraisal tools; ethical and regulatory guidance; implementation, workflow, and electronic medical record literature; and internal medicine-specific literature on complexity, longitudinal care, and transitions of care. These domains were selected because they correspond to recurrent governance functions required for safe AI use: transparent evaluation, critical appraisal, accountability, workflow integration, data stewardship, and clinical-contextual interpretation. The synthesis highlighted a persistent gap between cross-cutting AI governance instruments and the discipline-oriented governance needs of internal medicine. This gap does not imply that existing frameworks are inadequate, but that their principles require translation into departmental governance structures capable of addressing complex, longitudinal, and multimorbid care. In particular, existing frameworks insufficiently address the interaction between algorithmic tools and the clinical complexity of multimorbid patients, the care-continuum nature of internist work, the role of electronic medical records as both data sources and clinical interfaces, and the growing role of AI in mediating access to scientific evidence. To address this gap, we propose a five-pillar framework comprising: (1) clinical oversight and human-in-the-loop decision-making; (2) data integration oriented to clinical complexity; (3) organizational embedding across the care continuum; (4) ethical, legal, and regulatory governance; and (5) governance of scientific knowledge and AI-mediated evidence access. We also provide an operational checklist to support local implementation readiness. AI should not be introduced into internal medicine as an isolated technological layer, but should be governed as a complex clinical intervention. A discipline-oriented five-pillar model may help departments, hospitals, and scientific societies assess not only whether an AI tool performs well, but whether the surrounding clinical, organizational, data, regulatory, and epistemic conditions are mature enough to support safe use. © The Author(s), under exclusive licence to Società Italiana di Medicina Interna (SIMI) 2026.}, keywords = {Artificial intelligence; Clinical decision support systems; Clinical governance; Data governance; Digital health; Electronic medical records; Internal medicine; Multimorbidity}, } @article{aysanh2026300, title = {Artificial Intelligence as a Distributed Actor: Rethinking Organizational Theory Through Sociotechnical Networks}, author = {Aysan H.}, year = {2026}, journal = {Artificial Intelligence and Applications}, volume = {4}, number = {1}, pages = {92-100}, doi = {10.47852/bonviewAIA52025546}, abstract = {The rapid integration of Artificial Intelligence (AI) into organizational contexts is profoundly changing the fundamental assumptions in contemporary management and organizational theory. In response to these changes, this article introduces the Distributed Sociotechnical Agency Model (DSAM), a novel conceptual framework that synthesizes insights from Actor-Network Theory (ANT), sociomateriality, and principles of ethical governance. The DSAM framework conceptualizes AI as a distributed actor operating within complex networks and highlights three core dimensions: distributed agency, relational dynamics between human and non-human actors, and the imperative of ethical accountability. By using an integrated approach combining extensive literature review and analysis of real-life case studies, including a practical example involving an AI-driven university monitoring system, this study demonstrates how DSAM can facilitate more inclusive, user-friendly, flexible, responsive, and ethically aligned approaches to AI integration in organizations. To provide clarity and practical understanding, this article includes a detailed flowchart and a comparative table as well. Furthermore, recent academic studies and evolving policy developments are included in the discussion to strengthen the global relevance of the model. The article concludes by offering strategic recommendations for organizational governance and identifying promising directions for future empirical research. © The Author(s) 2025.}, keywords = {Actor-Network Theory; AI governance; distributed agency; organizational dynamics; sociomateriality}, } @article{prabhakrmabelbrisaevfmamadiyarovzraxmatullayevichkuerkinjontkhurramovam2026301, title = {A Multi-Layer Artificial Intelligence– Enabled Governance Framework for Integrating Information Security Management, Quality Management, and Strategic Human Resource Analytics: An Empirical Study on the Role of Corporate Social Responsibility in Sustainable Organizational Performance}, author = {Prabha K.R. and Mabel B.R. and Isaev F. and Mamadiyarov Z. and Raxmatullayevich K.U. and Erkinjon T. and Khurramova M.}, year = {2026}, journal = {Quality - Access to Success}, volume = {27}, number = {210}, pages = {418-426}, doi = {10.47750/QAS/27.210.45}, abstract = {This paper suggests a Multi-Layer Artificial Intelligence-facilitated Governance Framework that will combine Information Security Management, Quality Management, and Strategic Human Resource Analytics into a single decision-support system, as well as discuss the moderating role of Corporate Social Responsibility (CSR) on sustainable organizational performance. The study is based on the established management system standards, including ISO 9001, ISO 27001, and ISO 26000, and the conceptualization of the governance model aligning the strategic, tactical, and operational processes should improve the system reliability, transparency, and data-driven decision making. The proposed model is empirically tested by an empirical study of organisations of various industries based on structural equation modelling (SEM) showing that AI-enabled governance makes management systems alignment and interoperability significantly stronger and that CSR enhances the positive effect of integrated systems on the sustainability outcomes. The findings highlight the strategic significance of AI-driven governance to the management of the top management that aims at enhancing compliance, resilience, and long-term stakeholder value. Lastly, the research will provide some real-life solutions to the implementation of AI-improved governance structures since in addition to theoretical knowledge, the research will provide practical solutions to the organisations that seek excellence in terms of quality, security, and human capital performance. © 2026, SRAC-Romanian Society for Quality. All rights reserved.}, keywords = {AI Governance; CSR; Information Security Management; Quality Management; Sustainable Performance}, } @article{joshismkrishnansmps2026302, title = {Ethical implications of using AI for knowledge creation: a dialectic view from a meta-synthesis approach}, author = {Joshi S.M. and Krishnan S. and M.P. S.}, year = {2026}, journal = {Journal of Information, Communication and Ethics in Society}, pages = {1-18}, doi = {10.1108/JICES-12-2024-0190}, abstract = {Purpose – This paper aims to explain the ethical challenges of using Artificial Intelligence (AI) for knowledge creation. AI, endowed with autonomy and learning capabilities, strives to generate knowledge from data, thereby automating or augmenting knowledge work. Yet, reducing human involvement in knowledge-related activities may not always prove effective and can pose various ethical challenges to organizations. The authors adopt prevalent guidelines for ethical AI to reveal the problems organizations face when using AI in knowledge creation. Design/methodology/approach – The authors have analyzed the literature on knowledge generation using AI. This is followed by a meta-synthesis of qualitative studies to develop a comprehensive view of ethical issues arising out of the use of AI. Findings – The findings suggest multiple tensions between humans and AI during the training and continuation stages. The authors have used the dialectic lens to explain the ethical implications arising out of these tensions. Practical implications – AI promises to alter knowledge work fundamentally and is expected to benefit organizations and individuals. To ensure the highest level of ethical compliance, organizations must understand the inner mechanisms of using AI. This paper provides a comprehensive view to explain these mechanisms and also reveals the ethical issues that can emerge. Originality/value – The authors have taken an approach to identify ethical challenges when using AI in a real-world setting, which they believe is a first of its kind. Instead of simplified principle-based guidelines, they have synthesized the ethical challenges of using AI for knowledge creation. © 2026 Emerald Publishing Limited}, keywords = {AI ethics; Artificial intelligence; Knowledge management}, } @article{ahmadofmoriybretthauermdouradodahassancbisschopsrbhandaripbyrnemfdekkeremahadevanumayfpmessmannhmisawamogatahsaitoysilvermanalwangpyanotaabakkenlberzintm2026303, title = {The Legal and Ethical Framework for Artificial Intelligence in Gastrointestinal Endoscopy: A World Endoscopy Organization International Consensus Statement}, author = {Ahmad O.F. and Mori Y. and Bretthauer M. and Dourado D.A. and Hassan C. and Bisschops R. and Bhandari P. and Byrne M.F. and Dekker E. and Mahadevan U. and May F.P. and Messmann H. and Misawa M. and Ogata H. and Saito Y. and Silverman A.L. and Wang P. and Yano T. and Aabakken L. and Berzin T.M.}, year = {2026}, journal = {Annals of Internal Medicine}, volume = {179}, number = {2}, pages = {270-275}, doi = {10.7326/ANNALS-25-03415}, abstract = {The OperA (Optimising Colorectal Cancer Prevention through Personalized Treatment with Artificial Intelligence) project aims to transform colorectal cancer care through artificial intelligence (AI) innovations. Recognizing that legal and ethical challenges remain key obstacles to clinical integration, this Delphi study sought to identify and prioritize such concerns in the context of gastrointestinal (GI) endoscopy. Fourteen international experts participated in a 2-round Delphi process. In round 1, the steering committee, with feedback from participants, proposed legal and ethical issues pertaining to AI in endoscopy. Round 2 involved iterative rating and refinement of these issues to achieve consensus on their importance. Consensus was reached on 10 key statements spanning 3 thematic domains: data governance, medicolegal implications, and equity and bias. Experts emphasized the need for robust data protection, transparent algorithmic development, and institutional clarity on data ownership. Liability concerns related to AI-assisted diagnosis and automated reporting were highlighted, alongside calls for guidance from legal and professional bodies. Finally, participants underscored the importance of demographic diversity in training data sets and transparent reporting practices to mitigate bias and ensure equitable AI deployment. As AI tools become increasingly integrated into the clinical practice of gastroenterology, addressing legal, ethical, and equity-related challenges is essential. This expert consensus provides a foundation for developing guidelines and regulatory frameworks to support responsible AI adoption in GI endoscopy. © 2025 American College of Physicians.}, keywords = {Artificial Intelligence; Colorectal Neoplasms; Consensus; Delphi Technique; Endoscopy, Gastrointestinal; Humans; Liability, Legal; algorithm; Article; artificial intelligence; consensus; ethics; gastrointestinal endoscopy; human; legal aspect; medical liability; colorectal tumor; Delphi study; diagnosis; ethics; legal liability; prevention and control}, } @article{parkjakimchkimhj2026304, title = {AI literacy for safe deployment: Cross-national evidence on the interaction between talent and governance}, author = {Park J.-A. and Kim C.H. and Kim H.-J.}, year = {2026}, journal = {Telecommunications Policy}, volume = {50}, number = {1}, pages = {103106}, doi = {10.1016/j.telpol.2025.103106}, abstract = {This study investigates how national AI literacy and its operating environment jointly shape the frequency of AI-related incidents and hazards across countries. Drawing on cross-national panel data from 62 countries between 2014 and 2024, we integrate AI incident reports from the OECD AI Incidents Monitor with indicators from the Tortoise Global AI Index. AI literacy is proxied by the AI Talent Index, capturing the human capital available for AI development and deployment, while the AI Operating Environment Index reflects institutional and regulatory conditions supporting responsible AI use. Using a correlated random effects negative binomial model, we find that countries with higher AI literacy are associated with higher expected counts of reported AI-related incidents, consistent with greater exposure and capture. However, this relationship is significantly mitigated in countries with more mature operating environments. The interaction between AI talent and governance demonstrates a complementary risk-mitigation effect: in environments with robust safeguards, higher AI literacy leads to lower expected incident counts compared to environments with high literacy but weak governance. These findings suggest that building human capital without adequate governance may increase societal risk, and that effective AI policy must align investments in talent with regulatory infrastructure. Our results underscore the need for integrated national strategies to promote both capability and accountability in the era of rapidly advancing AI. © 2025}, keywords = {AI governance; AI literacy; AI safety; Correlated random effects; Cross-country analysis; Artificial intelligence; Investments; Personnel; AI governance; AI literacy; AI safety; Correlated random effect; Cross-country analysis; Cross-national; Human capitals; Operating environment; Panel data; Random effects; Random processes}, } @article{liurhany2026305, title = {Holding AI Accountable Like Herding Cats: The Contingent Impact on the Legitimacy of Algorithmic Bureaucracy; [“牧猫”之难下的AI问责制:对算法科层制合法性的权变影响]}, author = {Liu R. and Han Y.}, year = {2026}, journal = {Public Administration}, doi = {10.1111/padm.70043}, abstract = {The rise of artificial intelligence in public decision-making is reshaping state legitimacy by shifting administrative discretion from human bureaucracies to algorithmic systems. While research has explored AI accountability and legitimacy deficits, how they are related across different decision contexts remains unclear. Drawing on bureaucratic legitimacy, procedural fairness, and forum drifting theories, this study examines how AI accountability and effectiveness shape legitimacy perceptions, depending on decision outcomes. Using three survey experiments with 1135 participants in China, we find that accountability is most crucial when AI decisions introduce losses to citizens, whereas effectiveness plays a greater role when outcomes are positive to them. Additionally, the interaction effects between AI accountability and effectiveness are also contingent on decision outcomes. These findings advance AI governance research by highlighting the conditions under which algorithmic legitimacy is strengthened or weakened, emphasizing the need for tailored accountability and effectiveness strategies based on decision outcomes. © 2026 John Wiley & Sons Ltd.}, keywords = {accountability; algorithmic bureaucracy; decision outcome; effectiveness; legitimacy}, } @article{alimarimahmadw2026306, title = {From smart infrastructure to regenerative destinations: a tri-country study of tourism digital capabilities, innovation and ethical AI in Southeast Asia}, author = {Ali Mari M. and Ahmad W.}, year = {2026}, journal = {Tourism Review}, pages = {1-22}, doi = {10.1108/TR-11-2025-1289}, abstract = {Purpose – Tourism destinations increasingly pursue digital transformation, yet most initiatives remain efficiency-focused. Existing research provides a limited empirical explanation of how digital capability, environmental literacy and ethical artificial intelligence (AI) jointly enable regeneration beyond sustainability. To address this gap, this study aims to develop and test a smart regenerative tourism transformation model explaining how digital readiness, organizational capability and ethical governance support net-positive destination renewal. Design/methodology/approach – Survey data were collected from 543 tourism managers in Malaysia, Singapore and Thailand. Partial least squares structural equation modeling (PLS-SEM) was used to test eight hypothesized relationships linking smart tourism infrastructure, digital accessibility and inclusion, eco-literacy and net-zero commitment and regenerative destination governance to tourism digital transformation capability (TDTC), regenerative tourism innovation and regenerative destination transformation, with AI and data-ethics climate as a moderator. Findings – The results indicate that smart infrastructure, inclusivity and governance strengthen TDTC, which, in turn, supports regenerative tourism innovation and regenerative destination transformation. A strong AI and data-ethics climate amplifies these relationships. These findings are based on managerial perceptions and suggest, rather than confirm, destination-level regenerative progress. Research limitations/implications – This study’s cross-sectional design limits causal inference, as relationships remain correlational despite procedural and statistical checks. Future research should adopt longitudinal, experimental or panel data approaches to track TDTC over time. Additionally, incorporating objective indicators, such as AI ethics audits and digital investment records, can enhance validity. Expanding the sample to include diverse cultural contexts and adopting multi-stakeholder approaches will provide richer insights. Finally, dynamic-system modeling can better capture feedback loops between digitalization, governance and regeneration, advancing the understanding of regenerative tourism in evolving destinations. Practical implications – This study provides evidence for advancing regenerative digital transformation in Southeast Asian tourism. It confirms that smart tourism infrastructure, digital accessibility, eco-literacy and regenerative governance collectively enhance digital transformation outcomes. AI and data ethics climate play a crucial moderating role, ensuring ethical digital progress. Tourism boards must invest in inclusive digital ecosystems, ensuring that technology empowers local communities and businesses. Governments should integrate eco-literacy and sustainability into programs, while transparent AI frameworks should be adopted to ensure fairness. The findings support the creation of an Association of Southeast Asian Nations (ASEAN) regenerative tourism data network to align digital standards and promote sustainable, inclusive tourism. Social implications – This study highlights the role of digital transformation in fostering social inclusion and community well-being. By ensuring that technology enhances accessibility and empowers local communities, destinations can reduce socioeconomic inequalities and promote equitable growth. Regenerative governance frameworks and AI ethics are crucial for building trust and accountability in digital tourism, ensuring that innovations benefit all stakeholders. The integration of eco-literacy and sustainability practices further supports societal regeneration, encouraging active participation in conservation efforts and sustainable tourism practices. The proposed ASEAN regenerative tourism data network offers a model for fostering inclusive, responsible digital engagement across tourism destinations. Originality/value – This study offers a tri-country empirical examination of how ethical AI conditions the transformation of digital capability into regenerative value creation. It advances tourism theory by positioning digital transformation as a morally governed organizational capability that supports socio-ecological renewal rather than efficiency or harm reduction alone. © 2026 Emerald Publishing Limited}, keywords = {Digital transformation; Ethical artificial intelligence; Gobernanza inteligente; Inteligencia artificial ética; Regenerative tourism; Smart governance; Southeast Asia; Sudeste Asiático; Transformación digital; Turismo regenerativo; 东南亚; 伦理人工智能; 再生型旅游; 数字化转型; 智慧治理}, } @article{treinplemkenvaronef2026307, title = {Generalist and specialist roles of policy actors: Insights from preferences on AI regulation}, author = {Trein P. and Lemke N. and Varone F.}, year = {2026}, journal = {Public Policy and Administration}, doi = {10.1177/09520767251411201}, abstract = {This article examines how the roles of policy actors shape preferences for the regulation of artificial intelligence (AI). We distinguish between generalists, active across multiple policy sectors, and specialists, whose engagement is concentrated mostly in a single sector. We expect that generalists tend to support horizontal, cross-sectoral AI regulation, whereas specialists favor sector-specific, vertical approaches. Our analysis draws on original elite survey data comprising over 190 respondents from France, Germany, and Switzerland, including organizations and individuals engaged in AI governance across banking, health, and social welfare subsystems, as well as actors in the emerging AI policy sector. Using Bayesian regression models, we find indicative evidence that generalist actors—such as public interest groups, or trade unions—are more likely to endorse encompassing regulation, while specialists—such as occupational interest groups or public administrations—are more inclined toward sectoral regulation. This divergence becomes particularly pronounced when actors evaluate whether their own policy field should be integrated into a broader AI regulatory framework: policy fields that have many specialist actors, such as health policy, show markedly lower support for such integration. These findings contribute to public governance scholarship by clarifying how institutional roles influence regulatory preferences in the context of complex, cross-cutting policy challenges. The study contributes to our understanding of actor roles in policy subsystems and cross-sectoral policymaking as well as to Digital Era Governance in the public sector. © The Author(s) 2026. This article is distributed under the terms of the Creative Commons Attribution 4.0 License (https://creativecommons.org/licenses/by/4.0/) which permits any use, reproduction and distribution of the work without further permission provided the original work is attributed as specified on the SAGE and Open Access page (https://us.sagepub.com/en-us/nam/open-access-at-sage).}, keywords = {artificial intelligence (AI); cross-sectoral regulation; generalist versus specialist policy actors; policy integration; regulation; regulatory preferences}, } @article{lisjiafzhangs2026308, title = {Why Do Female Leaders Inhibit AI Adoption: The Role of Innovation Factors}, author = {Li S. and Jia F. and Zhang S.}, year = {2026}, journal = {IEEE Transactions on Engineering Management}, volume = {73}, pages = {2207-2221}, doi = {10.1109/TEM.2026.3661750}, abstract = {In recent years, the rapid advancement of artificial intelligence (AI) technology has profoundly transformed how businesses operate. Meanwhile, the role of female leaders in corporate decision-making is gaining more attention. Yet research on the impact of female leaders on the adoption of AI technology in companies remains limited. This study uses 32 601 firm/year observation data from A-share listed companies in China between 2007 and 2021 to explore the influence of female leaders on corporate AI adoption. Additionally, we examine the moderating effects of innovation-related factors, including innovation capability, innovation efficiency, and innovation quality, on this relationship. The study finds that the proportion of female leaders in boardrooms is negatively correlated with the adoption of AI technology in companies. This finding was validated by conducting five robustness tests and a placebo test and addressing endogeneity through the instrumental variable method. Furthermore, for firms that possess strong innovation capabilities, have a high level of innovation investment, produce abundant innovation output, and maintain high innovation quality, the inhibitory effect exerted by female leaders on AI adoption is significantly mitigated. This research has significant implications for AI ethics of engineering management literature identifying the risk aversion trait of female leaders in containing AI adoption and identifying the innovation factors affecting this relationship. © 1988-2012 IEEE.}, keywords = {Artificial intelligence (AI) adoption; female leaders (FLs); innovation factors; Artificial intelligence; Engineering research; Investments; Risk management; Artificial intelligence adoption; Artificial intelligence technologies; Corporates; Decisions makings; Female leader; Innovation capability; Innovation factor; Moderating effect; Observation data; Related factors; Decision making}, } @article{sohailssmadsendonadeemmcambriae2026309, title = {Ethical AI in Business}, author = {Sohail S.S. and Madsen D.O. and Nadeem M. and Cambria E.}, year = {2026}, journal = {IEEE Intelligent Systems}, doi = {10.1109/MIS.2026.3688455}, abstract = {Ethical risks posed by artificial intelligence (AI) in business drive a need for comprehensive frameworks that consider technically strong but human value-centric approaches. The current article proposes a novel conceptual framework grounded in three synergistic pillars: Attention, Intention and Intervention. Attention metaphorically represents an AI system’s awareness of ethical concerns such as bias, discrimination, and fairness. Intention emphasizes aligning AI objectives with business ethics and broader societal values from the design stage. Intervention highlights the critical role of human oversight and decision-making to guide and correct AI behavior. In addition, we have demonstrated through case studies that Attention, Intention and Intervention together can help developers, managers, and governance actors mitigate bias and foster trust. We compare the proposed triadic framework to existing ethical AI taxonomies and discuss its integration into corporate AI governance. We posit that the proposed interdisciplinary framework can serve as a guide for developing ethical AI in business. © 2001-2011 IEEE.}, } @article{raghupathiwsahariaakulkarnit2026310, title = {Identifying Key Issues in Artificial Intelligence Litigation: A Machine Learning Text Analytic Approach}, author = {Raghupathi W. and Saharia A. and Kulkarni T.}, year = {2026}, journal = {Applied Sciences (Switzerland)}, volume = {16}, number = {1}, pages = {235}, doi = {10.3390/app16010235}, abstract = {The rapid proliferation of artificial intelligence (AI) systems across high-stakes domains—with global AI adoption accelerating since 2023—has created an urgent need to identify which AI challenges and issues are materializing into real-world harms so that policymakers can develop targeted regulations, organizations can implement effective risk management, and accountability mechanisms can address actual rather than speculative problems. Public concern has risen sharply: 52% of Americans now feel more concerned than excited about AI (up from 38% in 2022), and 80% believe government should maintain AI safety rules even if development slows. Yet existing approaches exhibit critical limitations that impede evidence-based governance. Ethics frameworks, while establishing normative principles across 84+ published guidelines, remain aspirational rather than empirical. Survey-based studies capture perceptions from over 48,000 respondents globally but measure expectations rather than documented harms. Incident databases catalog over 1200 AI failures but depend on media coverage, systematically overrepresenting high-profile cases while underrepresenting routine organizational problems. This study addresses this gap by analyzing 347 AI-related U.S. litigation cases using machine learning text analytics, providing empirical evidence of AI problems that have crossed the threshold from abstract concern into documented legal conflict. Employing Latent Dirichlet Allocation (LDA) and Non-negative Matrix Factorization (NMF) topic modeling with coherence validation (NMF achieving 0.276 NPMI vs. LDA’s 0.164), the analysis identifies nine distinct AI issue areas with specific case distributions: cybersecurity vulnerabilities and data breaches (116 cases, 33.4%), intellectual property and AI ownership (61 cases, 17.6%), AI misrepresentation and inflated claims (59 cases, 17.0%), criminal justice and algorithmic due process (37 cases, 10.7%), employment automation (33 cases, 9.5%), privacy and surveillance (31 cases, 8.9%), platform accountability (21 cases, 6.1%), algorithmic bias (19 cases, 5.5%), and government AI deployment (6 cases, 1.7%). The findings reveal a systematic mismatch between AI ethics discourse—which emphasizes fairness and transparency—and litigation patterns, where data security (33.4%) and intellectual property (17.6%) dominate while algorithmic bias comprises only 5.5% of cases. Most disputes are addressed through existing legal frameworks (First Amendment, Lanham Act, FOIA, Title VII) rather than AI-specific regulation, underscoring the urgent need for governance mechanisms aligned with empirically documented AI challenges. © 2025 by the authors.}, keywords = {AI ethics; AI governance; AI litigation; algorithmic accountability; algorithmic bias; artificial intelligence; cybersecurity; intellectual property; LDA; legal text analysis; machine learning; NMF; text analytics; topic modeling; Abstracting; Automation; Crime; Cybersecurity; Data privacy; Learning systems; Machine learning; Product liability; Risk assessment; Risk management; Algorithmic accountability; Algorithmic bias; Algorithmics; Artificial intelligence ethic; Artificial intelligence governance; Artificial intelligence litigation; Cyber security; Latent Dirichlet allocation; Legal text analyze; Legal texts; Machine-learning; Matrix factorizations; Negative matrix factorization; Property; Text analysis; Text analytics; Topic Modeling; algorithm bias; article; artificial intelligence; clinical practice guideline; computer security; criminal justice; employment; epidemiology; evidence based practice; fairness; freedom of information act; government; human; information security; law suit; machine learning; non-negative matrix factorization; practice guideline; privacy; risk management; United States; vulnerability; Intellectual property}, } @article{roemmichkmartinkschaubf2026311, title = {CA–CI: Integrating Contextual Integrity and the Capabilities Approach for Dignity Considerations in AI Governance}, author = {Roemmich K. and Martin K. and Schaub F.}, year = {2026}, journal = {IEEE Security and Privacy}, doi = {10.1109/MSEC.2026.3654404}, abstract = {Capabilities approach -contextual integrity (CA–CI) extends contextual integrity through the integration of dignity thresholds from the capabilities approach and the specification of purpose as a constitutive parameter. We demonstrate how CA–CI can operationalize the EU AI Act's fundamental rights impact assessments, harm thresholds, and anticipatory governance. © 2003-2012 IEEE.}, keywords = {Artificial intelligence; Constitutive parameters; Contextual integrities; Impact assessments; Data integrity}, } @article{haritauganesanakrishnakumarishemasundarim2026312, title = {DYNAMIC CAPABILITIES AND ETHICAL STEWARDSHIP IN THE AGE OF AUTONOMOUS ORGANIZATIONAL DECISION MAKING}, author = {Harita U. and Ganesan A. and Krishnakumari S. and Hemasundari M.}, year = {2026}, journal = {Acta Innovations}, volume = {59}, pages = {301-307}, abstract = {Autonomous systems and artificial intelligence (AI) have changed the business model as know it by making decisions on behalf of the organization, making the business efficient, productive and responsive to the business strategy. The subsequent rise in the usage of autonomous technologies, however, introduced some ethical considerations in terms of transparency, accountability, biasness in the algorithms, privacy and control. The article has come up with a new theoretical framework which it has named the Dyn amic Ethical Capability Framework (DECF) which combines the ideas of the dynamic capability’s theory and the theory of ethical stewardship to assist in responsible AI-based decision making in firms. A quantitative research design, The Responsible AI Measures Dataset, is a resultant outcome that consists of more than 12,000 measures of ethical AI e valuation on equity, governance, explainability, responsibility and regulatory compliance. The validation of model and the relationship between the variables of study were done with the help of statistical and structural modeling packages such as SPSS. The results indicate that there is a strong positive relationship between ethical governance and organizational sustainability because the mean scores of three factors, ethical stewardship, adaptive governance capability, and sustainable organizational performance, are 4.42, 4.35, and 4.30, respectively. The results indicate that organizations with dynamic capabilities that are accompanied by ethical stewardship are able to significantly enhance autonomous decision quality, stakeholder trust, and long-term organizational resilience. The proposed framework plays its part in responsible AI governance by offering strategic guidance for responsible AI in autonomous organizational settings. © 2026 Rotherham Academic Press Ltd. All rights reserved.}, keywords = {artificial intelligence governance; autonomous decision making; dynamic capabilities; ethical stewardship; organizational trust; sustainable performance}, } @article{ofemo2026313, title = {Dynamic governance for agentic AI: synthesizing dual-use regimes through paradox theory}, author = {Ofem O.}, year = {2026}, journal = {AI and Society}, doi = {10.1007/s00146-026-03124-4}, abstract = {This study proposes a paradox-aware governance framework for the dual-use risks of agentic artificial intelligence. Anchored in paradox theory as the primary conceptual lens, the framework draws supplementary insights from institutional logics, dynamic capabilities, affordance theory, and panarchy to address context-specific implementation challenges. A critical-comparative analysis of historical dual-use regimes spanning nuclear, biological, chemical, cyber, and export controls yields five core governance paradoxes: Capability Control versus Innovation, Autonomous Action versus Accountability, Algorithmic Opacity versus Verification, Centralized Oversight versus Decentralized Development, and Proactive Safety versus Information Hazards. The study employs directed content analysis with transparent coding procedures, documenting both deductive category application and inductive emergence. Real-world cases including the GPT-4 deployment, Meta’s LLaMA model proliferation, and the EU AI Act’s tiered risk framework ground the propositions empirically. Each paradox is operationalized through paired mechanisms, boundary conditions, and governance levers such as Adaptive Capability Ledgers and Behavioral Compliance Sandboxes. The framework advances five falsifiable propositions specifying testable relationships between governance interventions and measurable outcomes. Equity is explicitly addressed through concrete instruments including compute-credit redistribution protocols, equity-weighted governance metrics, and capacity-building requirements for Global South stakeholders. Validation pathways encompassing quantitative indicators and qualitative stakeholder feedback ensure adaptive refinement. This integrated model transforms static regulations into living systems that co-evolve with AI’s inherent malleability, offering regulators, practitioners, and scholars actionable tools to sustain innovation while preempting irreversible harms. © The Author(s), under exclusive licence to Springer-Verlag London Ltd., part of Springer Nature 2026.}, keywords = {Adaptive regulation; Agentic artificial intelligence; AI governance framework; Dual-use paradox; Equity; Global South; Paradox theory; Technology control regimes; Adaptive control systems; Compliant mechanisms; Feedback; Laws and legislation; Living systems studies; Adaptive regulation; Agentic artificial intelligence; AI governance framework; Control regimes; Dual use; Dual-use paradox; Equity; Global south; Paradox theory; Technology control regime; Chemical hazards}, } @article{fantusslijwangttangl2026314, title = {Ethical Knowledge, Challenges, and Institutional Strategies Among Medical AI Developers and Researchers: Focus Group Study}, author = {Fantus S. and Li J. and Wang T. and Tang L.}, year = {2026}, journal = {Journal of Medical Internet Research}, volume = {28}, pages = {e79613}, doi = {10.2196/79613}, abstract = {Background: As artificial intelligence (AI) becomes increasingly embedded in clinical decision-making and preventive care, it is urgent to address ethical concerns such as bias, privacy, and transparency to protect clinician and patient populations. Although prior research has examined the perspectives of medical AI stakeholders, including clinicians, patients, and health system leaders, far less is known about how medical AI developers and researchers understand and engage with ethical challenges as they develop AI tools. This gap is consequential because developers’ ethical awareness, decision-making, and institutional environments influence how AI tools are conceptualized and deployed in practice. Thus, it is essential to understand how developers perceive these issues and what supports they identify as necessary for ethical AI development. Objective: The objectives of the study were twofold: (1) to examine medical AI developers’ and researchers’ knowledge, attitudes, and experiences with AI ethics; and (2) to identify recommendations to enhance and strengthen interpersonal and institutional ethics-focused training and support. Methods: We conducted 2 semistructured focus groups (60-90 minutes each) in 2024 with 13 AI developers and researchers affiliated with 5 US-based academic institutions. Participants’ work spanned a wide variety of medical AI applications, including Alzheimer disease prediction, clinical imaging, electronic health records analysis, digital health, counseling and behavioral health, and genotype–phenotype modeling. Focus groups were conducted via Microsoft Teams, recorded, and transcribed verbatim. We applied conventional qualitative content analysis to inductively identify emerging concepts, categories, and themes. Coding was performed independently by 3 researchers, with consensus reached through iterative team meetings. Results: The analysis identified four key themes: (1) AI ethics knowledge acquisition: participants reported learning about ethics informally through peer-reviewed literature, reviewer feedback, social media, and mentorship rather than through structured training; (2) ethical encounters: participants described recurring ethical challenges related to data bias, patient privacy, generative AI use, commercialization pressures, and a tendency for research environments to prioritize model accuracy over ethical reflection; (3) reflections on ethical implications: participants expressed concern about downstream effects on patient care and clinician autonomy, and model generalizability, noting that rapid technological innovation outpaces regulatory and evaluative processes; and (4) strategies to mitigate ethical concerns: recommendations included clearer institutional guidelines, ethics checklists, interdisciplinary collaboration, multi-institutional data sharing, enhanced institutional review board support, and the inclusion of bioethicists as members of the AI research team. Conclusions: Medical AI developers and researchers recognize significant ethical challenges in their work but lack structured training, resources, and institutional mechanisms to address them. Findings of this study underscore the need for institutions to consider embedding ethics into research processes through practical tools, mentorship, and interdisciplinary partnerships. Strengthening these supports is essential to preparing the next generation of developers to design and deploy ethical AI in health care. ©Sophia Fantus, Jinxu Li, Tianci Wang, Lu Tang. Originally published in the Journal of Medical Internet Research.}, keywords = {AI; AI ethics; artificial intelligence; ethics; focus group; medical; Artificial Intelligence; Female; Focus Groups; Humans; Male; Research Personnel; Alzheimer disease; Article; artificial intelligence-assisted technology; attitude; checklist; clinician; commercial phenomena; consensus; content analysis; counseling; diagnostic imaging; digital health; electronic health record; ethical compliance; feedback system; female; generative artificial intelligence; genotype phenotype correlation; human; information processing; institutional ethics; institutional review; knowledge; learning; machine prediction model; male; mentoring; patient care; peer review; personal experience; practice guideline; prediction; privacy; professional practice; qualitative analysis; scientist; social media; training; artificial intelligence; ethics; personnel}, }