Records

This joint report by UNESCO, the Inter-American Development Bank, and the OECD examines the effects of artificial intelligence on the working lives of women. It follows major stages of the workforce lifecycle, including job requirements, hiring, career progression, and upskilling, to address existing gender gaps. The publication explores whether the deployment of AI will help narrow gender inequality or perpetuate and widen it.

This academic article explores child-robot interaction in a real-life school environment to understand which features contribute to a trustworthy relationship and whether social robots can reduce stress in children. Through two experiments, the study found that most children trust a robot during their first interaction regardless of intonation, voice, or humor, though humor had a negative impact on trust for girls and children without prior robot experience. Additionally, both social robot interactions and classical music sessions lowered children's stress levels, though not significantly. The findings suggest potential for social robots to support children's well-being and help them make feelings explicit.

This empirical research study investigates the social and legal perceptions of embodied digital technologies, focusing on robots and users of prostheses. The authors conducted two studies evaluating how participants rate stimuli varying in prosthesis technicity and robot anthropomorphism across dimensions including competence, sociability, morality, legal personality, and decision-making authority. The findings reveal that technical sophistication influences competence and morality attributions for prostheses, while robot anthropomorphism affects perceptions of sociability, competence, legal personality, and decision-making authority.

ISO/IEC 38507:2022 provides guidance for members of the governing body of an organization to enable and govern the use of artificial intelligence. It helps ensure effective, efficient, and acceptable use of AI within public and private companies, government entities, and not-for-profit organizations. The document also provides guidance to executive managers, external specialists, public authorities, service providers, and auditors.

This strategic plan outlines the Equality and Human Rights Commission's priorities and regulatory approach for 2022 to 2025. It details six key areas of focus, including addressing the equality and human rights impact of digital services and artificial intelligence. The document sets out how the Commission will use its statutory powers, data, and evidence to protect rights and enforce equality law across Britain.

This NIST special publication outlines preliminary socio-technical guidance for identifying, understanding, managing, and reducing bias in artificial intelligence systems. It examines three categories of AI bias (systemic, statistical, and human) and discusses challenges across datasets, test, evaluation, validation, and verification (TEVV), and human factors. The document serves as a foundational step toward more detailed guidance to improve public trust and promote trustworthy and responsible AI.

This academic article examines the ethical challenges surrounding the use of robotic care assistants for the elderly, including issues related to privacy, autonomy, dignity, and loneliness. It explores how 'ethics by design' and institutional robot ethics committees can help ensure that robotic technology promotes user welfare and remains within moral bounds. The paper also reviews existing regulatory frameworks and highlights the complexities of informed consent in this domain.

This document outlines a research study protocol designed to explore patients' and healthcare professionals' expectations, attitudes, and the ethical acceptability of integrating socially assistive humanoid robots into nursing care at a regional clinical centre in Slovenia. The study utilizes a cross-sectional survey methodology incorporating previously validated questionnaires, including the Technology-Specific Expectation Scale, the Ethical Acceptability Scale, and the Negative Attitudes towards Robots Scale. The findings aim to inform implementation strategies and address potential misconceptions or fears regarding robotics in hospital workflows.

This research study examines the performance impact of a custom log engine and NoSQL database approach—both locally and in cloud environments—to support a log-based accountability and explainability system for autonomous robots. The authors propose a three-layer accountability architecture mapping software components, events, and robot behaviors to ensure traceability without breaking system performance. Through seven assessment scenarios using ROS and a simulated environment, the study evaluates CPU, network, disk, and message metrics to determine the most effective logging and storage strategies.

This academic paper examines the challenges, scenarios, and ethical considerations surrounding robotic speech synthesis in human-robot interactions. It highlights the technical difficulties of generating interaction-oriented speech such as backchannels, proposes a scenario classification framework for robot voice design, and discusses potential ethical issues including language, identity, gender, and aesthetic biases.

This academic paper examines the ethical implications of intentionally triggering gender biases through the design of virtually embodied conversational agents (ECAs). It reviews current literature on bias alignment and the feedback hypothesis, evaluates four distinct approaches ranging from full permissibility to proactive counter-stereotypical design, and concludes that a moderate approach avoiding discriminatory feedback is preferable. Ultimately, the authors discuss the complex trade-offs between interaction efficiency and moral responsibility in social robotics.

This viewpoint article discusses the indiscriminate use of surgical robots in medical practice and training, arguing that surgeons often apply robotic technology to simple procedures where it is neither necessary nor cost-effective. Citing the law of the instrument, the author cautions that robotic platforms should be properly targeted at complex procedures rather than routine surgeries like cholecystectomies. Furthermore, the article highlights concerns regarding resident training, increased healthcare costs, and underreported complication rates.

This academic article explores the ethics of social robotics by examining how embodiment and vulnerability differ between physical "true robots" and software-based "analytic agents". The author argues that an agent's mode of presence in the physical world conditions its capacity to learn and interact with humans, carrying major implications for the ethics and politics of artificial intelligence. By contrasting physical machines with disembodied bots and apps, the paper highlights how physical embodiment shapes vulnerability and moral agency.

This review paper examines the state of the art of robotic systems in agriculture, focusing on health, safety, ethics, and legislative issues. It notes that while agricultural robots offer economic and environmental benefits, they lack clear safety standards and a commonly accepted ethical code. The authors highlight emerging questions around ethical dilemmas, liability, privacy, personal data protection, and legislation, emphasizing the need for multidisciplinary cooperation.

This academic article offers a conceptual framework for AI robot accountability that combines descriptive and normative ethical theories, specifically integrating the locus of morality and moral intensity. The authors introduce four 'accountability clusters' ranging from professional norms to supra-territorial regulations to map how different AI robot application contexts interact with ethical categories. By moving beyond a simple binary view of ethics, the framework categorizes AI robot uses into illegal, immoral, permissible, and supererogatory behaviors.

This academic paper offers a critical analysis of the prevailing ethical approach to social assistive robotics (SAR). It identifies three major shortcomings in current debates: an individual-centered perspective, a narrowed understanding of ethical concepts like freedom and responsibility, and a lack of discussion on SAR teleology. To address these gaps, the authors propose a comprehensive ethical framework grounded in applied ethics, political philosophy, and care ethics.

This academic article defends an internalist and event-causal libertarian view of free will and responsibility, arguing that advanced non-conscious robots can possess a genuine capacity for moral responsibility despite lacking phenomenal consciousness. By combining a developmental self-theory with a consequentialist framework of holding agents responsible, the author outlines how machines can learn from feedback and alter their behavior autonomously over time. Finally, the paper addresses common objections regarding design history, responsibility gaps, and self-modification capabilities in autonomous systems.

This academic paper examines the ethical dilemmas surrounding the application of autism spectrum disorder (ASD) nursing robots, focusing on responsibility, patient morality, emotional deception, and privacy. The author argues that robots cannot fully replace physiotherapists and should remain in an auxiliary, supervised role. Complete substitution by nursing robots is currently found to be unfeasible due to safety, moral, and emotional risks.

This academic paper examines liabilities arising from adversarial cyberattacks on surgical robots from both a Danish national and European Union law perspective. It proposes a taxonomy for adversarial and non-adversarial failures in cyber-physical medical systems and evaluates how existing frameworks like the EU Medical Device Regulation and Danish tort and product liability apply. The authors also discuss administrative remedies such as the Danish Patient Compensation system as a streamlined mechanism for patient recovery.

The Governance Guidelines for Implementation of AI Principles provide action targets and practical examples to assist companies in designing and implementing AI governance systems based on human-centric AI principles. The document covers conditions and risk analysis, goal setting, system design, implementation, evaluation, and re-analysis within an agile governance framework. It is designed as a living document to support voluntary corporate efforts and foster multi-stakeholder understanding.

This academic study investigates how ethical issues in human-robot interaction influence users' intention to use service robots in frontline delivery contexts. Using structural equation modeling on 341 survey responses, the research evaluates six ethical dimensions: social cues, trust and safety, autonomy, responsibility, privacy and data protection, and labor replacement. The findings reveal that trust and safety, social cues, privacy protection, and responsibility significantly impact user intention, whereas autonomy and replacement concerns do not directly influence usage intention.

The National Artificial Intelligence Action Plan for Thailand (2022–2027) sets a comprehensive strategic roadmap to develop the country's AI ecosystem, enhance workforce skills, and drive economic and social impact. It outlines five core strategies focusing on AI ethics and regulations, infrastructure development, talent acquisition and education, technology and innovation, and cross-sector adoption. The plan aims to position Thailand among the global leaders in AI readiness by improving governance, expanding digital infrastructure, and integrating AI into key industries like healthcare and agriculture.

This academic article examines the implementation and use of feeding robots for people with impaired arm function, drawing on an ethnographic study conducted in Denmark and Sweden. It analyzes how policy strategies promoting self-reliance, design visions of empowerment, and everyday care scripts entwine. The study argues that care robots generate alternative choice-dependency situations rather than straightforward self-reliance or obsolescence of care providers.

This academic article explores how artificial intelligence and extended reality technologies challenge traditional legal and regulatory frameworks by enabling the direct design and engineering of human behavior and perception. The author argues that AI and XR can manipulate internal decision-making and external sensory experiences, effectively creating regulatees that embody desired outcomes rather than complying with external norms. This form of regulatory incorporation paradoxically mirrors individual agency, rendering contemporary legal doctrines ill-equipped to recognize or address the resulting harms.

This EU-OSHA report examines the implications of advanced robotics and automation for occupational safety and health (OSH). It uses a comprehensive taxonomy to evaluate the effects of full and semi-automation across various tasks and economic sectors, detailing psychosocial, physical, and organisational impacts. Finally, it outlines relevant standards, risk assessment challenges, and provides recommendations for human-centred workplace applications.

This policy brief examines the occupational safety and health implications of advanced robotics and automation, focusing on human-robot interaction, anthropomorphic design, system transparency, and worker trust. It highlights how design choices influence workers' expectations, safety, and well-being, while recommending the application of established ergonomic principles and appropriate transparency to foster effective collaboration. Commissioned by EU-OSHA, the brief provides insights for designing safer and more intuitive automated workplaces.

This policy brief explores the occupational safety and health (OSH) risks and opportunities associated with advanced robotics and automation in the workplace. It analyzes human-robot interaction across four main dimensions: function allocation, task design, interaction design, and operation and supervision. The document provides recommendations on worker involvement, upskilling, and adherence to principles such as 'human in control' and transparency during robotic implementation.

This report by the European Agency for Safety and Health at Work (EU-OSHA) sets the scene for its four-year research programme on digitalisation, examining types and definitions of AI-based systems and advanced robotics for task automation. It maps current and potential uses across cognitive and physical task domains, reviews sectoral distributions, and provides an overview of relevant EU and national policies, strategies, programmes, gaps, and needs in relation to occupational safety and health (OSH).

This World Health Organization report examines how ageism manifests within artificial intelligence applications designed for health care. It analyses potential biases against older populations in AI-driven health technologies and offers guidance on ensuring age-inclusive design and deployment. The document provides recommendations for stakeholders across the health and technology sectors to prevent discrimination and promote equitable health outcomes for older adults.

This document establishes the Artificial Intelligence Readiness and Maturity Assessment Framework for Rwanda to monitor the progress of its National AI Policy. Developed by MINICT in cooperation with C4IR and GIZ, the framework uses a data-driven tool supported by a dynamic dashboard to assess readiness across six priority dimensions. It provides a detailed analysis of the first-round assessment, highlighting current capabilities, data gaps, and recommendations to strengthen Rwanda's AI ecosystem.

This document outlines an artificial intelligence policy roadmap for the Caribbean region, developed by the UNESCO Office for the Caribbean. It details UNESCO's ongoing work and upcoming initiatives regarding AI standards, capacity building, research, and regional consultations. The roadmap aims to raise awareness of AI, identify regional priorities, and leverage digital transformation for sustainable development.

This chapter examines the risks, challenges, and dark side of artificial intelligence and robotics within the context of the Fourth Industrial Revolution. It proposes an integrated responsible innovation framework that incorporates anticipation, inclusion, reflexivity, and responsiveness as a science governance mechanism. The study emphasizes that organizations and stakeholders must share collective responsibility and establish ground rules such as trust, transparency, and co-responsibility for sustainable technological innovations.

This academic paper proposes a methodology for care robots to learn ethical rules of behavior from experience under the supervision of a human ethical teacher. It utilizes a combination of Answer Set Programming (ASP) for ethical knowledge representation and Inductive Logic Programming (ILP) to generate logical rules from scenarios. Case studies illustrate how robots can learn specific rules regarding patient safety and pandemic restrictions to evaluate future actions and ensure transparency.

This academic paper proposes a refined code of ethics specifically targeted at robotic engineers, drawing upon a review of existing literature and applying the guidelines to a surgery case study. The code covers areas such as engineer responsibility, human supervision, transparency, default ethical programming for robots, patient rights, and professional conduct. It emphasizes that while engineers hold responsibility for their creations and client selection, operational liability often shifts to external buyers or supervising professionals like surgeons.

This report examines the impact of cognitive automation and artificial intelligence on occupational safety and health (OSH). It utilizes a systematic literature review methodology to categorize AI-based systems into full and semi-automation tasks across person-related, information-related, and object-related domains. Furthermore, it identifies key OSH implications, focusing heavily on psychosocial effects such as workload, job transformation, deskilling, trust, loss of autonomy, and privacy concerns.

This book chapter presents an empirical study investigating the social and ethical ramifications of robotic elderly care from the perspective of both care recipients and care providers. Through generative phenomenographic interviews with 36 actors, including older adults and informal and formal caregivers, the study uncovers expectations, latent fears, and ethical concerns surrounding care robots. The findings highlight a multi-actor perspective captured through thematic maps on assistance, monitoring, and companionship, showing that while robots could improve well-being, caregivers remain somewhat reluctant to accept them as co-providers.

This research article investigates elders' expectations and experiences with a companion-type social robot named Pleo through an empirical study involving 33 participants. The study assesses both anticipated and fulfilled capabilities, as well as the impact on elders living independently or with assistance in community or nursing home environments. Findings reveal that while positive therapeutic and enjoyable effects were noted, negative outcomes such as undue responsibility, fading enjoyment, or anger and frustration also occurred.

This consensus statement provides ethical and practical guidance on developing and deploying narrow AI applications for robotic surgical training curricula. Utilizing a three-round Delphi process with an international panel of 30 clinical, academic, and industry experts, the statement addresses key areas including data protection and privacy, reproducibility and transparency, predictive analytics, inherent biases, and accountability. The guidance aims to establish standardized approaches to utilizing surgical data while ensuring ethical deployment and patient safety.

This document serves as the introductory overview to a comprehensive six-part set of guidelines for safe collaborative robot design and implementation in workplaces. It defines key terminology related to collaborative robots (cobots), systems, and workplaces, while categorising physical, psychological, and ethical hazards. Additionally, it outlines five guiding principles for creating a cobot-safe workplace and summarizes safety measures across the technology's lifecycle.

This paper by the IEEE Global Initiative on Ethics of Autonomous and Intelligent Systems provides guidance for businesses and policymakers on incorporating well-being metrics—such as ESG standards and UN Sustainable Development Goals (SDGs)—into the design and lifecycle of artificial intelligence systems (AIS). It advocates for 'Well-being by Design' as a core metric for responsible AI to prioritize human well-being and environmental sustainability over solely financial indicators. The document outlines practical steps for assessing, applying, and scaling these metrics across organizations of varying sizes.

This academic book chapter explores ethical design considerations in robotics, examining the non-neutrality of technology and how normative ethical theories apply to engineering practices. It introduces various ethics tools such as checklists, standards, design principles, and ethical frameworks, alongside the Value Sensitive Design (VSD) methodology. The chapter concludes with a case study on the value-sensitive design of a Danish healthcare drone and discusses responsible research and innovation frameworks.

This working paper develops a dynamic general equilibrium growth model to analyze how robotization, which replaces unskilled labor, interacts with fiscal and monetary policies. The authors find that while robotization increases economic growth and the skill premium, implementing targeted fiscal and monetary policies can mitigate wage inequality at the cost of an efficiency-equity trade-off. Overall, the presence of fiscal and monetary policies leads to higher long-run economic growth compared to a scenario without policies.

This qualitative research study explores the technological risks and ethical implications associated with the adoption of robots in long-term care homes through semi-structured interviews with thirty participants. The findings highlight key challenges such as safety concerns, increased staff workload, privacy violations, cost, social justice issues, and the tension between human connection and technology. To address these issues, the authors propose a set of practical recommendations centered on the ETHICS framework: Engagement of stakeholders, Technology benefit and risk assessment, Harm mitigation, Individual autonomy, Cultural safety and justice, and Support of privacy.

The National AI Policy for the Republic of Rwanda serves as a strategic roadmap to leverage artificial intelligence for economic growth, improve quality of life, and position the country as a global and regional innovation hub. It establishes six priority policy areas covering 21st-century skills, infrastructure, data strategy, public sector adoption, private sector growth, and practical ethical guidelines. Furthermore, it outlines concrete recommendations, sector-specific deep dives, and an implementation plan to ensure responsible and inclusive AI development across Rwanda.

The National AI Policy of the Republic of Rwanda serves as a comprehensive roadmap to harness artificial intelligence for sustainable economic growth while mitigating associated risks. Developed by MINICT and RURA with international partners, the policy outlines six priority areas ranging from 21st-century skills and reliable infrastructure to robust data strategies and practical ethical guidelines. It aims to position Rwanda as a leading African innovation hub and center of excellence in AI through concrete implementation plans and public-private collaboration.

This academic paper describes the design, implementation, and assessment of a virtue ethics education module integrated into a senior-level Mobile Robotics technical elective course. The module uses a character education framework based on the Jubilee Centre model to cultivate virtues such as empathy, intellectual humility, and practical wisdom among engineering students. Through weekly discussions, student-led artifacts, and a final role-play case study, the module aims to improve student efficacy in ethical reasoning and reflection on professional qualities in AI and robotics.

These provisions regulate the use of algorithmic recommendation technology in internet information services within mainland China. They establish requirements for service providers to uphold mainstream values, protect user rights, manage information security, and comply with filing and security assessment obligations. The regulations are issued jointly by multiple Chinese state departments to promote healthy industry development and safeguard national security and public interests.

This academic paper investigates the proposal to create a legal electronic personhood for AI-equipped robots, drawing on the European Parliament resolution with recommendations on Civil Law on Robotics. It highlights the risks and problems of this initiative, arguing that it relies on anthropomorphic rhetoric and a pragmatic corporate model without addressing the underlying legal grammar and embodiment issues. The author concludes that electronic personhood obscures accountability and that ethical considerations and differentiated liability mechanisms should precede status definitions.

This study presents a storytelling social robot controlled via the ACT-R cognitive architecture that utilizes persuasive techniques and ethical stances to converse about COVID-19 rules and vaccines. The model integrates a needs-driven approach and cognitive theories to evaluate dialogue management, storytelling strategies, and argumentative techniques. An exploratory evaluation involving 63 participants assessed the effectiveness of these persuasive strategies and ethical settings.

This paper develops an ethical and political framework for introducing responsible AI and robotics to healthcare based on the perspective of Science and Technology Studies and the ethics of care. The authors propose moving away from polarized utopian or dystopian narratives towards managing uncertainties through three movements: shifting focus to uncertainties, replacing human-robot interaction with a 'Robot Embedded in a Network' approach, and shifting from general normative frameworks to contexts of use. Ultimately, they argue for practical, progressive actions termed 'caring in the in-between' involving relationship monitoring, stakeholder inclusion, and making fears and hopes commensurable.