Records

The first legally binding international treaty on AI (CETS No. 225). Parties must adopt measures so that activities across the AI lifecycle are consistent with human rights, democracy and the rule of law - including transparency and oversight, equality, privacy, remedies for violations, and a risk and impact management framework. Opened for signature in Vilnius in September 2024; in force since November 2025, open to accession by non-member states.

The Voluntary AI Safety Standard provides practical guidance to all Australian organisations on how to safely and responsibly use and innovate with artificial intelligence. Developed by the National Artificial Intelligence Centre, the standard consists of 10 voluntary guardrails that apply across the AI supply chain to help mitigate and manage risks. It aligns with international standards to ensure businesses complying with it also meet international requirements.

UNESCO has launched two new AI competency frameworks, one for students and one for teachers, designed to guide countries in integrating artificial intelligence into education safely and ethically. The student framework focuses on a human-centred mindset, ethics, techniques, and system design, while the teacher framework covers professional development, AI foundations, ethics, and pedagogy. These frameworks serve as a roadmap to ensure AI empowers learners and educators while respecting human rights and cultural diversity.

The UAE's International Stance on Artificial Intelligence Policy outlines the country's strategy for shaping global AI governance, building international partnerships, and promoting responsible AI development. It establishes six core guiding principles: advancement, collaboration, community, ethics, sustainability, and safety. The document aims to position the UAE as a global AI leader while driving economic diversification and improving quality of life both domestically and internationally.

This public policy document outlines the United Arab Emirates' international stance on artificial intelligence policy, emphasizing global leadership, ethical governance, and strategic partnerships. It establishes key guidelines around progress, collaboration, societal inclusion, ethics, sustainability, and safety to ensure responsible AI development. The policy serves as a strategic extension of the UAE's domestic AI framework, aiming to drive economic diversification and improve quality of life locally and globally.

This policy brief explores the transformative potential of artificial intelligence for youth empowerment and sustainable development across Africa. It examines the current e-readiness and digital landscape of the continent, highlighting opportunities in agriculture, healthcare, education, small and medium-sized enterprises, and digital trade. Finally, it outlines actionable recommendations for policymakers, governments, and stakeholders to foster an enabling environment for AI-driven youth development.

This study examines the challenges of embedding responsible innovation (RI) into research and development practices within a three-year project developing a socially assistive robot. Using participatory action research (PAR), the authors fostered collective learning and identified four key barriers: lack of an action perspective, the noncommittal nature of RI, the misconception that co-design equals RI, and limited integration between R&D task groups. The paper concludes with recommendations for R&D teams and funding bodies to better institutionalize RI.

The Garis Panduan Tadbir Urus dan Etika Kecerdasan Buatan (AIGE), launched in September 2024 by MOSTI, provides strategic guidance for the safe, ethical, and responsible development and use of artificial intelligence in Malaysia. Operating under the theme "AI for Malaysia, AI for Semua," the guidelines aim to strengthen public trust, mitigate ethical risks, and enhance technology adoption across multiple sectors while remaining voluntary. The framework supports the National Artificial Intelligence Roadmap 2021-2025 and several United Nations Sustainable Development Goals across key national application domains.

The United Nations Global Digital Compact establishes an international framework for digital cooperation aimed at closing digital divides, expanding digital economic inclusion, protecting human rights online, ensuring responsible data governance, and enhancing international AI governance. It outlines concrete objectives, cross-cutting principles, and commitments for UN member states and multi-stakeholder partners to be achieved by 2030. Key governance mechanisms include the creation of an Independent International Scientific Panel on AI and a Global Dialogue on AI Governance within the United Nations.

This final report by the UN High-Level Advisory Body on Artificial Intelligence outlines the urgent need for a holistic, global approach to governing AI for humanity. It identifies critical governance gaps regarding representation, coordination, and implementation on an international scale. To address these challenges, the report recommends enhancing global cooperation through mechanisms like an international scientific panel on AI, regular policy dialogues, standards exchanges, and a dedicated UN AI office.

The Riyadh Charter on Artificial Intelligence in the Islamic World, established by ICESCO, provides a dynamic and strategic foresight-based procedural framework to govern the ethical use of AI across member states. It integrates cultural and Islamic values into AI development, focusing specifically on ICESCO's core competencies of education, science, and culture. The charter outlines core principles for AI systems, humans, and institutions, and offers actionable recommendations to address regional disparities and promote responsible innovation.

This perspective article examines the ethical, legal, and regulatory challenges surrounding the integration of artificial intelligence, automation, and surgical robots in healthcare. Drawing historical parallels from the widespread adoption of early surgical robots, it highlights issues related to patient safety, informed consent, training deficiencies, and the liability dilemma of apportioning blame in autonomous human-robot systems. The author argues for proactive regulations, standards like the Society of Automotive Engineers driving automation taxonomy, and robust training programs to prevent repeating past mistakes as surgical automation advances.

This document introduces Part III of an academic volume focusing on narrative approaches to human-robot interaction and the law. It examines how legal narratives, metaphors, and storytelling shape the understanding, liability, and regulation of emerging technologies such as autonomous vehicles and trading robots. The text discusses specific research chapters covering Norwegian trading robot cases, Singaporean newspaper coverage of autonomous vehicles, and the need for interdisciplinary hermeneutics.

This legislative bill establishes a comprehensive legal framework for the development, implementation, and use of artificial intelligence in Panama, aiming to protect human rights, privacy, and safety while fostering technological innovation. It introduces a risk-based classification system for AI applications, outlines the obligations of developers and providers, establishes a National Artificial Intelligence Policy, and creates a High-Level AI Commission under SENACYT. Ultimately, the bill was rejected and defeated during its first committee debate in April 2026.

The National Artificial Intelligence Strategy 2024-2027 of Costa Rica establishes a roadmap to promote the ethical, secure, and responsible use, adoption, and development of artificial intelligence. It sets forth guiding principles centered on human dignity, transparency, equity, and sustainability while outlining strategic actions across seven thematic axes. The strategy seeks to maximize socioeconomic benefits, modernize public administration, foster regional innovation, and proactively manage technical and ethical risks.

This academic article explores the legal, ethical, and game-theoretic framework for operating autonomous delivery robots (ADRs) in pedestrian areas. It addresses the 'freezing robot problem' by proposing that ADRs be legally classified as pedestrians and permitted to make credible threats of inflicting small penalties (such as personal space invasion) based on principles of robot self-defence and the defence of property. The authors examine existing UK traffic regulations, product liability, negligence laws, and data protection implications under the GDPR.

This paper introduces dynamic certification as an adaptive regulatory and governance framework for responsible embodied AI systems. It argues that traditional static certification mechanisms fail to capture the fluid, evolving, and context-dependent nature of AI behaviors in real-world environments. The proposed approach combines formal modeling, early-stage simulation, transitional controlled testing, and confirmatory post-deployment monitoring to incrementally expand operational envelopes while maintaining ethical standards and societal values.

This report presents the findings of Wave 4 of the Public Attitudes to Data and AI tracker survey, conducted by Savanta on behalf of the UK’s Responsible Technology Adoption Unit. It examines public awareness, trust, and sentiment regarding data sharing and artificial intelligence across over 5,000 UK adults. The survey highlights growing public trust in data practices alongside persistent concerns about security, misinformation, and the societal impact of AI.

This qualitative study explores the perceptions of professional nursing personnel regarding potential ethical risks in intelligent robotic health education. Conducting semi-structured interviews with 17 nursing professionals in China using a descriptive phenomenological approach, the study identifies six main themes of ethical concern: threats to human dignity, patient safety, privacy disclosure, implicit burdens, responsibility attribution, and expectations for social support. The findings emphasize the need to carefully consider and address the potential negative impacts of health education robots in clinical practice.

The Continental Artificial Intelligence Strategy outlines an Africa-centric, development-focused approach to artificial intelligence to support Agenda 2063 and Sustainable Development Goals. It was endorsed by the African Union Executive Council during its 45th Ordinary Session in July 2024 to promote ethical, responsible, and equitable AI practices. The strategy calls for unified national approaches among AU member states to position Africa as a leader in inclusive AI development.

The Italian Strategy for Artificial Intelligence 2024-2026 sets out Italy's national plan to foster AI development and adoption while ensuring alignment with European values and regulations. Structured around four macro-areas—Research, Public Administration, Enterprises, and Training—the strategy aims to enhance national competitiveness, improve public services, support economic growth, and address skills gaps through coordinated investments and governance mechanisms. It emphasizes the creation of national knowledge assets, public-private partnerships, and robust monitoring to mitigate risks such as non-action, homogenization, and digital divide.

The Artificial Intelligence Act (Regulation (EU) 2024/1689) is the world's first comprehensive legal framework for artificial intelligence, establishing a risk-based approach to regulate AI systems within the European Union. It introduces prohibitions on specific unacceptable-risk AI practices, strict compliance obligations for high-risk AI applications, and transparency rules for general-purpose AI models and generative content. The regulation is designed to promote trustworthy AI while protecting fundamental rights, safety, and security across the EU market.

Nigeria's National Artificial Intelligence Strategy outlines a comprehensive framework to position the country as a global leader in responsible, ethical, and inclusive AI innovation. The document sets out strategic pillars covering infrastructure, ecosystem building, sector adoption, ethical development, and governance. It provides a roadmap for leveraging AI to drive economic growth, social development, and technological advancement while addressing local challenges such as infrastructure limitations and workforce skills.

This page provides an overview of the Artificial Intelligence Act (Regulation (EU) 2024/1689), the world's first comprehensive legal framework on artificial intelligence. It details the risk-based approach categorizing AI systems into unacceptable risk, high risk, transparency risk, and minimal or no risk, alongside rules for general-purpose AI models. The legislation sets out timelines for implementation, governance through the European AI Office, and compliance support tools.

This report by the U.S. Copyright Office addresses the legal and policy issues related to artificial intelligence and copyright, focusing in Part 1 on digital replicas. It analyzes existing state and federal legal frameworks, concluding they are insufficient to address the rapid rise of AI-generated deepfakes and voice clones. Consequently, the Office recommends the enactment of new federal legislation to protect all individuals from unauthorized digital replicas during their lifetimes, alongside specific licensing guardrails, OSP safe harbors, and First Amendment accommodations.

This document outlines a legislative bill for artificial intelligence in Colombia, establishing a comprehensive legal framework to ensure ethical, responsible, competitive, and innovative development. It proposes an adaptive normative approach focused on risk management, fundamental rights protection, and institutional capability building. Key measures include risk-based classification of AI systems, institutional governance led by MinCiencias, regulatory sandboxes, labor protections, and ethical data traceability.

This resolution by the Argentine Ministry of Security establishes the Artificial Intelligence Applied to Security Unit (UIAAS). The unit is tasked with leveraging artificial intelligence for the prevention, detection, investigation, and prosecution of crime within the jurisdiction of federal security forces. Its functions include social media monitoring, facial recognition, predictive policing, and threat detection.

This document is a cross-sectoral profile and companion resource to the AI Risk Management Framework (AI RMF 1.0) specifically for Generative Artificial Intelligence. It defines risks unique to or exacerbated by generative AI, such as confabulation, data privacy impacts, and CBRN information risks. Additionally, it provides a set of suggested actions and risk management strategies aligned with trustworthy AI characteristics.

This document augments the secure software development practices and tasks defined in the Secure Software Development Framework (SSDF) version 1.1 by adding recommendations, considerations, notes, and informative references specific to generative AI and dual-use foundation model development. Developed to support Executive Order 14110, the profile provides guidance for AI model producers, AI system producers, and AI system acquirers. It covers risks across the software and AI model development life cycle, such as data poisoning, training data provenance, model weight protection, and continuous monitoring.

The Italian Strategy for Artificial Intelligence 2024-2026 sets out a comprehensive national roadmap to foster AI research, public administration efficiency, enterprise innovation, and education. Issued by the Agency for Digital Italy, the strategy prioritizes anthropocentric, reliable, and sustainable AI solutions tailored to Italy's industrial and social context. It addresses key risks such as technological non-action, cultural homogenization, over-regulation, and the digital divide through targeted governance, data infrastructure, and funding mechanisms.

This research report explores the opportunities and challenges of closer human-robot interaction in the workplace, drawing on survey data and case studies investigating advanced robotic systems. It examines the impacts of automation on work organisation, job profiles, productivity, and working conditions, including psychosocial health and safety risks. The publication contributes to the broader policy debate on the automation of work by highlighting the need for human-centric design and active worker involvement.

This joint ILO and World Bank working paper estimates the exposure of Latin American and Caribbean labor markets to generative AI by leveraging harmonized household and labor force surveys. It adjusts occupational exposure measures to account for the slower pace of technology adoption and the digital divide, evaluating whether digital infrastructure barriers prevent workers from realizing potential productivity gains. The findings indicate that urban jobs requiring higher education and formal status face higher exposure, while lack of digital access significantly limits potential augmentation benefits across the region.

The African Union Continental AI Strategy establishes a comprehensive framework to harness artificial intelligence for Africa's socio-economic transformation, cultural renaissance, and sustainable development. Guided by principles of ethics, human rights, and inclusion, it focuses on maximizing AI benefits across key sectors, building necessary infrastructure and capabilities, minimizing risks, and fostering regional and international cooperation. The strategy outlines fifteen specific action areas and calls upon member states, the African Union Commission, the private sector, and development partners to collaborate in implementing its imperatives.

The African Union Continental Artificial Intelligence Strategy outlines a comprehensive framework to harness AI for socio-economic development, cultural renaissance, and the achievement of Agenda 2063 goals. It emphasizes a people-centric, inclusive, and ethically grounded approach focused on maximizing AI benefits, mitigating risks, building local infrastructure and skills, and fostering regional and international cooperation. The strategy was adopted by the Executive Council of the African Union in July 2024.

This article examines the medicolegal challenges and legal implications surrounding the growing use of robotic surgery. It discusses the history and categories of surgical robots, advantages and disadvantages, training standards, surgical responsibilities, and informed consent requirements. Additionally, the paper analyzes legal liability, malpractice litigation data, and real-world case examples of surgical complications related to robotic systems.

The National AI Strategy Roadmap 2.0 (NAISR 2.0) is the Philippines' Department of Trade and Industry framework for driving an AI-enabled economy and inclusive industrial transformation. Launched in July 2024 as an update to the 2021 roadmap, it incorporates generative AI developments and ethical governance. The strategy has since been superseded by a new national AI strategy under the Department of Science and Technology.

This academic commentary responds to critiques of the author's previously proposed eco-relational approach in robot ethics. It addresses methodological differences concerning ethical concepts versus systems and explores the transition toward an ecologically responsible framework that accounts for non-human entities. The author emphasizes the need to bridge philosophical relational approaches with applied ethics and law rather than maintaining a separation.

This research report provides an evidence-based understanding of current opportunities and challenges for using artificial intelligence to improve the efficiency and service delivery of police corporate and enabling functions across Five Eyes countries. It examines AI technologies used in documentation, audio and video analysis, data integration, and biometrics, alongside call management, community interaction, and recruitment. The report identifies four main operational challenges—performance, integration and governance, data privacy, and internal trust—and outlines strategic opportunities and framework alignments for New Zealand Police.

This academic study examines the civil liability implications of artificial intelligence and smart robots within the Jordanian legal framework. Utilizing doctrinal legal research, the authors identify significant gaps in Jordan's current laws regarding AI-induced liabilities, jurisdictional ambiguities, and the limitations of traditional personal and objective liability theories. The paper proposes several reforms, including legislative intervention, the potential creation of a legal personality for AI, global accountability standards, and international cooperation.

This document is Amazon's submission to the Canadian Competition Bureau's consultation on artificial intelligence and competition. It outlines Amazon's activities in AI development, discusses the state of competition and innovation within the AI sector, and highlights Amazon's commitment to the responsible management and use of AI technologies.

This document is Amazon's submission in response to the Canadian Competition Bureau's request for feedback on artificial intelligence and competition. It outlines Amazon's activities in AI, details how competition and innovation are working to the benefit of customers, and highlights Amazon's commitment to the responsible development and use of AI. The paper advocates for the applicability of traditional competition laws rather than burdensome sector-specific regulations to ensure innovation continues to flourish.

This document is a proposed Artificial Intelligence Law introduced in the Grand National Assembly of Turkey (TBMM) in June 2024. The draft legislation establishes a comprehensive regulatory framework for AI systems, covering definitions, core ethical principles, risk management, conformity assessments, and enforcement mechanisms. It outlines obligations for AI providers, distributors, users, importers, and operators, while setting substantial administrative fines for violations.

The National Framework for the Assurance of Artificial Intelligence in Government establishes a nationally consistent approach for Australian, state, and territory governments to safely and responsibly adopt AI. Grounded in Australia's AI Ethics Principles, the framework provides cornerstones and practical implementation guidance covering governance, data governance, risk-based approaches, standards, and procurement. It aims to build public trust and confidence while ensuring the lawful, ethical, and transparent use of AI across public service delivery.

This academic research article proposes the Principle of Alternative Possibilities (PAP) as an ethical research framework for evaluating human sexual interactions with robots. The author analyzes three main scenarios involving robot development, distribution, and engagement, arguing that social agents remain morally accountable for their choices because alternative possibilities are available. The framework is conceptual and pre-experimental, relying on a qualitative review of existing philosophical and academic literature.

This paper presents EHAZOP, a method for conducting ethical hazard analyses on assistive robots by adapting collaborative safety engineering techniques. The authors describe a proof-of-concept case study applying EHAZOP to the Ari assistive robot across cognitive, social, and coaching functions. Results demonstrate that EHAZOP successfully identifies diverse ethical hazards, including novel concerns not captured in standard frameworks like BS8611.

This regulation establishes a comprehensive, risk-based legal framework for artificial intelligence across the European Union, aiming to improve the functioning of the internal market while promoting human-centric and trustworthy AI. It prohibits certain unacceptable AI practices, sets strict mandatory requirements and obligations for high-risk AI systems, and introduces transparency rules for specific AI systems and general-purpose AI models. Additionally, the regulation supports innovation through regulatory sandboxes and governance structures at both national and EU levels.

This regulation establishes a uniform legal framework for artificial intelligence across the European Union, promoting human-centric and trustworthy AI while protecting health, safety, and fundamental rights. It sets harmonized rules for the placing on the market, putting into service, and use of AI systems, addressing risks associated with high-risk applications and prohibiting certain harmful practices. Additionally, the regulation supports innovation and ensures the free movement of AI-based goods and services across the internal market.

This regulation establishes a comprehensive legal framework for artificial intelligence in the European Union, introducing harmonized rules across Member States. It classifies AI systems based on a risk-based approach, prohibiting certain unacceptable AI practices, imposing strict requirements on high-risk AI systems, and laying down transparency rules for specific AI applications. Additionally, it establishes governance structures at both the national and Union levels, including the European Artificial Intelligence Board and the AI Office, while supporting innovation through regulatory sandboxes.

This research paper presents an empirical design justice study to identify ethical considerations at the intersection of large language models (LLMs) and social robotics. The author conducted a lab-based co-design and interaction study with nine participants to examine how the physical embodiment of social robots can perpetuate or escalate ethical issues found in LLMs. The findings map ethical considerations across four conceptual dimensions: interaction, co-design, terms of service, and relationship. Ultimately, the study evaluates the design justice methodology as an approach to incorporate diverse communities and address structural power dynamics in emerging AI technologies.

The UAE Charter for the Development and Use of Artificial Intelligence outlines national objectives, priorities, and expected outcomes to establish the UAE as a global hub for AI. It focuses on ethical and responsible AI development, human oversight, data privacy, safety, and compliance with local laws and international treaties. The charter addresses multiple stakeholders across the public and private sectors to ensure AI serves the public good and enhances quality of life.