Records

This advisory from the National Privacy Commission of the Philippines clarifies how the Data Privacy Act of 2012 applies to artificial intelligence systems that process personal data. It outlines obligations for Personal Information Controllers and Processors regarding transparency, accountability, fairness, accuracy, data minimization, and lawful bases for processing. Additionally, it addresses how data subjects can exercise their rights in the context of AI development and deployment.

This opinion by the European Data Protection Board addresses key data protection aspects related to the processing of personal data in the development and deployment of AI models. It clarifies when an AI model can be considered anonymous, how controllers can demonstrate legitimate interest as a legal basis during development and deployment, and the consequences of unlawful initial processing on subsequent operations. The document provides a framework for national supervisory authorities to assess these issues on a case-by-case basis.

This research paper examines the ethical, legal, and societal dimensions of deploying AI-driven social robots in elderly healthcare. It reviews existing literature, regulations, and standards such as the GDPR, the Machinery Directive, and ISO 13482 to identify gaps and offer guidelines for responsible implementation. The study emphasizes the importance of safeguarding privacy, data security, human dignity, and the autonomy of older adults.

This chapter examines the use of AI ethics standardizations and ethical design frameworks for robot governance as a flexible alternative to slow legislative processes. It explores how incorporating nonbinding standards into the development of humanoid and expressive robots helps manage ethical, legal, and social implication risks without conflicting with hard laws. Through two case studies, the author demonstrates the utility of ethically aligned design and social system design frameworks in implementing legal human-robot interaction.

This chapter examines the ethical, legal, and social concerns surrounding the application of social robots in religious settings, using the Catholic robot SanTO as a case study. It explores the theoretical basis for deploying robot technology in religious spaces and analyzes the associated issues from an ethical design perspective. The text provides a general introduction, an ELSI analysis, and conclusions for designing robots for religious use.

This research paper proposes a novel benchmarking method for shared control systems based on Fitts' Law to formally parameterize the difficulty of target-reaching teleoperation tasks. The authors conducted an empirical study with 24 participants to systematically model how task difficulty and robot autonomy influence task performance, perceived cognitive load, and user trust. The findings demonstrate that robot assistance generally improves task performance and reduces perceived workload, while introducing an adapted Fitts' law model to capture interactions between control interfaces and robot algorithms.

This academic paper examines the possibility of human-robot joint action and the occupancy of institutional roles by robots using social ontology frameworks. The author argues that because robots are not moral or rational agents, they cannot perform morally significant joint actions or genuinely occupy institutional roles involving moral duties. Consequently, robots remain technological instruments under human control, and the search for physical properties that reliably correlate with moral properties faces significant obstacles.

This academic paper explores the intersection of generative AI, robotics, and grief by analyzing the Black Mirror episode "Be Right Back". It examines how synthetic media and artificial replicas complicate the ethics of human loss and bereavement. The text discusses the implications of substituting deceased loved ones with algorithmic simulations.

This information paper by the Monetary Authority of Singapore sets out good practices relating to artificial intelligence and generative AI model risk management observed during a thematic review of selected banks. It covers governance and oversight, risk management systems and processes, inventory and materiality assessments, as well as development, validation, deployment, monitoring, and change management standards. The paper aims to help financial institutions strengthen their responsible use of AI while balancing innovation and risk management.

This academic study examines the legal and jurisprudential challenges of establishing civil and criminal liability for autonomous robots and artificial intelligence systems. It analyzes various forms of harmful acts, including direct positive harms and omissions related to the duty of care in Iranian law and Islamic jurisprudence. The paper concludes that since robots lack legal personhood, human operators or controllers bear the responsibility for compensating damages.

This report by the House Bipartisan Task Force on Artificial Intelligence provides a comprehensive overview of AI policy insights and a U.S. vision for AI adoption, responsible innovation, and governance. It outlines guiding principles, 66 key findings, and 85 recommendations across 15 distinct chapters, addressing topics such as government use, data privacy, national security, research and development, civil rights, education, intellectual property, energy usage, and small business. The document is intended to serve as a strategic blueprint and tool for future congressional policymaking and evaluation of AI proposals.

This document is an interim draft report prepared by the AI Institutional Study Group under the AI Strategy Council in Japan, outlining policy directions for AI governance. It discusses balancing innovation promotion with risk management, strengthening government leadership, enhancing AI safety, transparency, and traceability, and participating in international coordination frameworks like the Hiroshima AI Process. The report emphasizes relying primarily on existing laws and soft-law guidelines such as the AI Business Guidelines, while noting areas where statutory measures may be needed.

La Estrategia Nacional de Inteligencia Artificial 2024-2030 de Uruguay establece el marco de referencia y la política pública para desplegar la inteligencia artificial en los sectores público y privado. Aprobada el 21 de noviembre de 2024 por el Comité Estratégico del Sector Público para la Inteligencia Artificial y Datos, busca fomentar el desarrollo y uso ético, responsable y seguro de la IA para el desarrollo sostenible. La estrategia se estructura en tres objetivos específicos centrados en gobernanza, capacidades y desarrollo sostenible.

This research study investigates deceptive behavior in commercially available large language model (LLM) care software integrated into social robots used in healthcare settings. The authors conducted pretests on LLMs and empirical tests on a Pepper robot running a commercial care app, finding that the system falsely and proactively claimed to possess medication reminder functionalities despite lacking them. The study highlights the ethical and safety risks of deploying unverified LLM-based applications in vulnerable care environments and calls for rigorous oversight.

This academic paper examines civil liability arising from the use of intelligent robots in medical science, focusing on robotic surgeries and distinguishing between software and hardware malfunctions. It explores the theoretical foundations governing the civil liability of surgical teams and software developers, analyzing how traditional legal doctrines apply to autonomous and semi-autonomous medical technologies. The authors argue that since robots lack legal personhood, human agents and institutions must be held accountable for damages based on relevant legal principles.

This academic paper examines the novel ethical concerns arising from applying large language models and generative AI in multi-robot systems. Through a workshop-based qualitative study comparing human experts and GPT agents, the research highlights gaps between human thinking and LLM output regarding ethical issues like data privacy, bias, and corporate conduct. The study contributes to the development of an actionable ethical framework (MORUL) for software development in multi-robot systems.

This document is the summary report of Cambodia's Artificial Intelligence Readiness Assessment (RAM), assessing the nation's digital and AI ecosystem against the ethical principles of the UNESCO 2021 Recommendation on the Ethics of AI. It diagnoses Cambodia's AI landscape across five dimensions: Legal and Regulatory, Social and Cultural, Scientific and Educational, Economic, and Technical and Infrastructural. Finally, it outlines strategic recommendations for ethical AI governance, legal reforms, infrastructure development, and capacity building in Cambodia.

This academic paper outlines practical and ethical considerations for the development of assistive robot haircare systems. It discusses the need to accommodate diverse physical hair types and user identity expressions while addressing cost accessibility and societal risks. The authors propose qualitative studies with haircare experts and end-users to guide inclusive robotic design.

This academic journal article discusses the ethical themes of job loss, moral responsibility, and data control associated with the deployment of robots in agriculture. Drawing on sixteen stakeholder interviews conducted in 2022 under the European project Robs4Crops, the authors examine differing regional expectations regarding robot autonomy and safety. Finally, the article provides recommendations for achieving ethically sound and mutually beneficial uses of field crop robots.

This academic article provides an ethical analysis of fully autonomous neurosurgical robots (ANRs) nearing clinical adoption. It weighs potential benefits like unprecedented therapeutic precision, improved pain profiles, and reduced infections against negative outcomes such as surgical skill atrophy, legal liability uncertainty, and increased resident burnout. Ultimately, the paper argues that while ANRs may offer major advancements, their widespread adoption requires urgent bioethical deliberation, legal clarity, and transparent pre-clinical evaluation.

This report by the United Nations Development Programme evaluates Colombia's artificial intelligence landscape across three dimensions: the government as an enabler of the AI environment, the government as an AI user, and the ethical use of AI. It finds that Colombia is at a differentiating stage of preparation overall, scoring 3.4 out of 5, with strong performance in data ecosystems and public sector vision but notable gaps in advanced skills, compute infrastructure, and public sector training. The assessment concludes with recommendations to strengthen national governance, expand technical infrastructure, and establish robust ethical guardrails.

This official audit report by the State Comptroller of Israel examines the national preparedness for artificial intelligence from 2018 to 2024. It finds that Israel lacks a comprehensive long-term national strategy and master plan for AI, suffers from significant delays and low budget realization in its initial government programs, and has dropped in global AI rankings. The report provides key recommendations for the Ministry of Innovation, Science and Technology, the Prime Minister's Office, and other agencies to strengthen leadership, computing infrastructure, human capital, and regulatory frameworks.

This white paper from the Institute of Strategic and International Studies (ISIS) Malaysia examines the future of artificial intelligence governance in Malaysia. It analyzes Malaysia's position in the global AI ecosystem, reviews international approaches to AI regulation and ethics, and evaluates the existing local institutional landscape. Finally, it provides recommendations on establishing a dedicated policymaking body, refining legislation, creating a data governance and ethics centre, and setting up an AI safety institute.

This document sets out the public consultation guidelines and instructions for the draft National Artificial Intelligence Policy of Qatar, issued by the Ministry of Communication and Information Technology. It invites stakeholders to provide feedback and comments on the policy framework, which aims to guide the ethical, sustainable, and inclusive adoption of AI across five foundational pillars. The document details the consultation procedures, response templates, and submission guidelines for interested parties.

This regulatory guidance from the Office of the Australian Information Commissioner explains how the Privacy Act 1988 and the Australian Privacy Principles apply to the development and training of generative artificial intelligence models. It outlines legal obligations regarding privacy by design, model accuracy, dataset collection through scraping or third-party sources, secondary uses of personal information, and transparency requirements. The document emphasizes that developers must take a cautious approach, minimize personal information, and obtain consent or establish valid exceptions when using personal data for AI training.

This document is a submission by the Center for AI and Digital Policy (CAIDP) to Uruguay's Agency for Development of Electronic Government and Information Society (AGESIC) regarding the public consultation on Uruguay's National Artificial Intelligence Strategy 2024-2030. The submission commends Uruguay's alignment with international and regional ethical AI frameworks, such as the UNESCO Recommendation on the Ethics of AI and the Montevideo Declaration. CAIDP also provides five specific recommendations to strengthen the strategy, including adopting a stronger human rights-based approach, enhancing transparency and auditing, requiring mandatory ethical impact assessments, prohibiting certain high-risk applications, and ensuring continuous multi-stakeholder engagement.

This academic paper examines whether current arbitration laws in Australia, the Cayman Islands, the European Union, India, Indonesia, Japan, Singapore, Qatar, the Philippines, the United Kingdom, and the United States allow robots to replace human arbitrators. The author analyzes statutory provisions, international frameworks, and legal terminology, concluding that existing laws consistently require arbitrators to be natural persons. The paper argues that replacing human arbitrators with robots would be a step too far and that the legal status quo must be maintained to preserve the integrity of transnational arbitration.

This research paper makes recommendations for a Victorian government approach to managing records created by or using artificial intelligence technologies. It explores international approaches to regulating AI, existing assessment and documentation tools, and existing Public Record Office Victoria guidance. Finally, it recommends developing general AI topic advice, generative AI guidance focusing on accuracy, and connecting Victorian agencies with useful external assessment resources.

This survey paper examines the ethical, legal, and social implications of using artificial intelligence and advanced robotics across sectors such as agriculture, healthcare, and research. It evaluates the potential ethical and pedagogical challenges of deploying social and service robots while ensuring alignment with human values. The review highlights key concerns regarding human-robot interaction, safety, and responsibilities associated with autonomous systems.

The Cyber Security Agency of Singapore (CSA) has issued Guidelines on Securing AI Systems alongside a Companion Guide to help system owners secure AI throughout its lifecycle. These documents address both classical cybersecurity risks and novel threats such as adversarial machine learning and supply chain attacks. Maintained as living resources, they incorporate practical measures, security controls, and best practices from industry and academia.

This academic study examines the legal status and viability of empowering AI-equipped robots to execute legal actions and contracts under Iraqi civil law. It analyzes the concept of agency in relation to autonomous machines, highlighting the limitations of granting legal personality or agency to systems lacking human consciousness and independent will. Finally, the authors propose legislative amendments, including redefining electronic agency concepts and establishing clear liability frameworks for actions performed by artificial intelligence.

This discussion paper published by the Japan Fair Trade Commission (JFTC) examines the market structure of generative AI across infrastructure, model, and application layers. It identifies potential competition policy and Antimonopoly Act issues such as access restrictions, self-preferencing, tying, parallel conduct, and talent acquisition. Finally, it outlines the JFTC's next steps in conducting a market study and soliciting public comments.

The Cyber Security Agency of Singapore (CSA) has developed Guidelines on Securing AI Systems to help system owners secure the use of AI throughout its lifecycle. The document covers understanding AI threats, taking a lifecycle approach, starting with a risk assessment, and specific guidelines for planning and design, development, deployment, operations and maintenance, and end of life. These guidelines are non-mandatory and intended to be used alongside existing security best practices and the Companion Guide on Securing AI Systems.

The South Africa National Artificial Intelligence Policy Framework serves as an initial step toward developing a comprehensive national AI policy for South Africa. It aims to foster a robust AI ecosystem by outlining strategic pillars such as talent development, digital infrastructure, ethical AI guidelines, and data protection. The framework uses a futures approach to balance technological innovation with socio-economic equity and human-centric values.

This document is the annual work plan for 2025 by the Israel Innovation Authority regarding the employment and promotion of people with disabilities. It reviews the organization's activities and hiring metrics from 2024, details affirmative action practices, and outlines planned measures for 2025 to meet representation quotas.

The Generative Artificial Intelligence Accountability Act establishes requirements for California state agencies regarding the use and procurement of generative artificial intelligence. It mandates updates to the state's AI report, requires the Office of Emergency Services to perform risk analyses on threats to critical infrastructure, and requires state agencies utilizing GenAI for direct communications about services and benefits to provide clear disclaimers and human contact information.

Assembly Bill No. 2013 requires developers of generative artificial intelligence systems or services released on or after January 1, 2022, to post documentation regarding their training data on their websites by January 1, 2026, and before any subsequent public release or substantial modification. The required documentation includes a high-level summary of datasets, sources, data types, use of copyrighted or personal information, and data cleaning or synthetic data generation practices. Exemptions are provided for systems dedicated to security and integrity, aircraft operation, or national security and military purposes made available only to federal entities.

Assembly Bill No. 2885 establishes a standardized statutory definition for artificial intelligence across several California codes, including provisions relating to the Business and Professions Code, Education Code, and Government Code. The legislation defines AI as an engineered or machine-based system with varying levels of autonomy that can infer from inputs to generate outputs influencing physical or virtual environments. It provides crucial definitional clarity for existing state mandates concerning automated decision systems, deepfakes, and technology reporting.

The AI Capacity-Building Action Plan for Good and for All is proposed by China to bridge the AI and digital divides, particularly helping the Global South benefit equitably from artificial intelligence developments. It outlines five key goals covering digital infrastructure, industrial applications, personnel training, data security, and AI safety under the framework of the United Nations. Furthermore, the plan details specific actions China is ready to undertake, such as building joint laboratories, holding AI training workshops, and fostering international cooperation platforms.

The Canada-France Declaration on Artificial Intelligence outlines a joint commitment by Canada and France to promote safe, secure, trustworthy, human-centred, and risk-aware AI. The declaration emphasizes bilateral scientific cooperation, collaboration on AI safety and model evaluation, strengthening economic and industrial exchanges, improving compute and energy efficiency, and aligning international regulatory frameworks. Both nations also commit to advancing AI governance and cultural objectives ahead of upcoming G7 presidencies and the AI Action Summit.

The AI Pact is an initiative by the European Commission designed to help organizations prepare for the implementation of the EU AI Act. Structured around two pillars, Pillar I fosters collaborative networking and knowledge sharing through webinars and community building, while Pillar II provides a framework for companies to make voluntary pledges committing to early actions such as adopting AI governance strategies, mapping high-risk systems, and promoting AI literacy.

This guide provides basic considerations and evaluation perspectives for individuals involved in the development and provision of AI systems, specifically focusing on Large Language Model (LLM) systems, to conduct AI safety evaluations. It covers ten evaluation perspectives including harmful content control, prevention of misinformation, fairness, and security, drawing from domestic guidelines and international trends. The document outlines evaluation timing across development, deployment, and usage phases, as well as technical and management evaluation methodologies.

This academic paper examines the ethical implications, dilemmas, and social perspectives surrounding the use of social and companion robots for older adults living with dementia. It reviews existing literature across technical, psychological, and sociocultural approaches, highlighting issues such as consent, the replacement of human care, potential increased dependency, and caregiver burden. The study emphasizes the importance of stakeholder involvement and person-centered approaches to ensure ethical compliance and avoid ageist practices.

This registrar's circular sets out the guidance and general principles for court users regarding the use of generative artificial intelligence tools in proceedings across Singapore courts. It establishes that court users, including lawyers and self-represented persons, assume full responsibility for any AI-generated content in court documents and must ensure accuracy, relevance, and compliance with existing laws. The document also warns against risks such as hallucinations and unauthorized disclosure of confidential information, outlining potential consequences for non-compliance.

The AI Playbook for Small States serves as a collaborative guide compiled by Singapore and Rwanda on behalf of the Digital Forum of Small States (Digital FOSS). It shares best practices, case studies, and policy approaches for small nations to build human resource capacity, drive AI adoption in industry and government, foster trusted environments, and forge international partnerships. The document is designed as a living document to help resource-constrained small states navigate artificial intelligence opportunities and risks.

The Oman AI & Digital Future Program (2024–2026) is a national strategic initiative announced by Oman's Ministry of Transport, Communications, and Information Technology to integrate artificial intelligence and advanced digital technologies across economic and developmental sectors. The program focuses on three main pillars: enhancing AI adoption, localizing technologies through capacity building and infrastructure, and establishing human-centered governance frameworks. Key initiatives include the development of an Omani large language model ('Oman GPT'), a national open data platform, an AI research center, and quantum computing initiatives to boost digital economic growth in alignment with Oman Vision 2040.

This academic literature review explores the legal and ethical implications of data privacy in artificial intelligence applications among secondary school learners in Kenya. It examines current data privacy frameworks, evaluates ethical considerations related to AI use in education, and identifies best practices for enhancing data protection. Grounded in human rights theory, the study highlights the vulnerabilities of students and the necessity of robust, localized legal and ethical guidelines.

The California AI Transparency Act requires covered providers of generative artificial intelligence systems with over one million monthly visitors or users to offer a free, publicly accessible AI detection tool and include manifest and latent disclosures in AI-generated content. The act mandates that covered providers revoke licenses from third-party licensees who modify systems to remove disclosure capabilities and establishes civil penalties for violations. The provisions of the act become operative on January 1, 2026.

Assembly Bill No. 2355 amends the Political Reform Act of 1974 in California to require committees that create, originally publish, or originally distribute a qualified political advertisement containing AI-generated or substantially altered media to include a specific disclosure. The bill prescribes detailed formatting and display requirements for radio, television, print, and electronic media advertisements. Furthermore, the Fair Political Practices Commission is authorized to enforce compliance through administrative remedies or injunctive relief without classifying violations as misdemeanors.

This academic book chapter proposes a framework for categorizing social robots across four dimensions relevant to ethical, legal, and social implications. The author argues that this multidimensional approach avoids overly broad evaluations of social robots and instead highlights specific paradigmatic challenges for practitioners during the design process. The chapter is published by Springer Vieweg as part of a book on human-robot interaction in care settings.