Records

This user manual outlines the functions and operation of the National Science and Technology Council's shared regulations database system. It explains how to navigate database scopes such as laws, administrative regulations, and draft rules, and describes search, viewing, and public consultation features. The manual aims to assist users in efficiently retrieving regulatory information and maintaining database quality.

This document outlines the proposed basic act for artificial intelligence in Taiwan, aiming to establish fundamental principles and guidelines for AI development and governance. It provides a legal framework to address societal impacts, rights protection, and promotion of AI technologies. The legislation seeks to balance innovation with ethical and safety considerations.

This legislative bill establishes a comprehensive framework to promote the research, development, and utilization of artificial intelligence-related technologies in Japan. It outlines basic principles, mandates the formulation of a basic AI plan, and sets up an Artificial Intelligence Strategic Headquarters within the Cabinet to ensure coordinated governmental policy and foster international cooperation.

This parliamentary report analyzes the implementation and budget allocation of Taiwan's AI Action Plan 2.0 for the 114th fiscal year, noting a planned expenditure of over 15.7 billion NTD. It identifies weaknesses in cross-departmental coordination and highlights the need to establish a comprehensive review mechanism, advance AI legal frameworks, and address surging power and talent demands. Ultimately, the report urges the government to strengthen sub-project linkages to fortify Taiwan's national AI competitiveness.

This legislative research report examines the product liability challenges posed by autonomous smart robots due to their machine learning and decision-making capabilities. It reviews current Taiwanese laws such as the Civil Code and Consumer Protection Act, discusses international developments including the European Union's Civil Law Rules on Robotics and the proposed AI Liability Directive, and analyzes the difficulties in assigning responsibility, causation, and legal personhood. Finally, it suggests balancing consumer protection with supply chain responsibilities and considering legal reforms or dedicated legislation for smart robots in Taiwan.

This academic chapter examines the intersection of ergonomics, ethical values, and robot ethics, focusing on how socio-technical systems and frameworks like Value Sensitive Design (VSD) can be applied to robotics. It highlights the integration of ethical considerations in technology development, drawing examples from care robotics and nursing home assistants. Finally, it outlines meta-ethical perspectives concerning robots as potential moral agents.

The UNESCO Ethical Impact Assessment (EIA) is a practical evaluation tool designed to help governments, companies, and researchers assess the ethical implications and alignment of individual AI systems with UNESCO's Recommendation on the Ethics of Artificial Intelligence. It provides a structured, six-stage process involving cross-functional teams, independent review, collaborative workshops, and stakeholder engagement to identify positive and negative impacts across the AI lifecycle. The tool is supported in part by the European Union and aims to promote responsible, transparent, and ethical AI development and deployment.

This article examines the legal boundary between medical malpractice and product liability as surgical robots gain increasing autonomy. It introduces the Autonomy-Control-Risk Attribution (ACRA) framework to help determine when liability should remain centered on medical malpractice, shift toward product liability, or be shared among clinicians, healthcare institutions, and manufacturers. The framework evaluates the autonomous function involved, meaningful human control, risk provenance, informational control, and interaction-based failures to prevent surgeons from becoming a moral crumple zone.

A mesterséges intelligenciára vonatkozó stratégia alkalmazása (Apply AI Strategy) az EU átfogó mesterségesintelligencia-ágazati stratégiája, amelynek célja a versenyképesség növelése, az innováció ösztönzése és a technológiai szuverenitás erősítése az EU-ban. A stratégia támogatja az első a mesterséges intelligencia politikát, fellendíti az MI elterjedését a tíz kulcsfontosságú ipari ágazatban és a közszférában, valamint új irányítási rendszert, az Apply AI Alliance-t hozza létre. Emellett intézkedéseket irányoz elő a készségek fejlesztésére, a digitális innovációs központok átalakítására és az európai adatokra épülő biztonságos modellek előmozdítására.

The Apply AI Strategy is the EU's overarching artificial intelligence sectoral strategy designed to enhance the competitiveness of strategic sectors, strengthen technological sovereignty, and boost AI adoption among SMEs and the public sector. It features sectoral flagships across 10 key industry sectors, support measures to tackle cross-cutting challenges, and a new governance framework including the Apply AI Alliance and AI Observatory. The strategy is complemented by related initiatives such as the AI in Science strategy, the Data Union Strategy, and the Cloud and AI Development Act.

Die Strategie „KI anwenden“ ist die übergreifende sektorale Strategie der EU für künstliche Intelligenz und soll die Wettbewerbsfähigkeit strategischer Sektoren sowie die technologische Souveränität stärken. Sie fördert die Einführung und Innovation von KI in ganz Europa, insbesondere bei kleinen und mittleren Unternehmen, und legt den Schwerpunkt auf eine „KI an erster Stelle“-Politik sowie einen „buy European“-Ansatz. Die Strategie umfasst branchenspezifische Leitinitiativen für zehn Schlüsselbranchen und den öffentlichen Sektor, Maßnahmen für eine KI-fähige Belegschaft und ein neues Governance-System wie die Apply AI Alliance.

The AI Application Strategy is an overarching EU strategy aimed at boosting the competitiveness of strategic sectors, strengthening technological sovereignty, and promoting AI adoption and innovation across Europe, particularly in SMEs. It introduces a 'AI-first' policy, a 'Buy European' principle in the public sector, and includes sector-specific flagship initiatives, support measures for a skilled workforce, and a new governance framework such as the AI Application Alliance. The strategy is complemented by related initiatives including the AI in Science strategy, the Data Union strategy, and the Cloud and AI Development Act (CADA).

This article examines the complex medico-legal challenges and liability frameworks associated with pediatric robotic-assisted surgery, analyzing a ten-year institutional experience alongside a narrative literature review. The authors highlight that robotic surgery shifts liability from an exclusively operator-fault model to a shared network involving the surgeon, healthcare institution, and device manufacturer. The paper discusses technical incidents encountered during procedures, the regulatory void surrounding algorithmic and product liability, and the necessity of establishing multi-stakeholder accountability frameworks.

The Apply AI Strategy is the European Union's comprehensive sectoral strategy aimed at increasing the competitiveness of strategic sectors and reinforcing EU technological sovereignty. It seeks to promote AI adoption and innovation across Europe, particularly among small and medium-sized enterprises (SMEs), through sectoral flagship initiatives, supporting measures for technological sovereignty, and a new governance system. The strategy is complemented by related initiatives such as the Artificial Intelligence in Science strategy and the Data Union strategy.

The Apply AI Strategy is a comprehensive EU sector-specific strategy aimed at improving the competitiveness of strategic sectors, strengthening Europe's technological sovereignty, and promoting AI adoption and innovation, particularly among small and medium-sized enterprises (SMEs). It introduces a "prioritise AI" policy, sectoral flagship initiatives across ten key industries and the public sector, and a new governance structure like the Apply AI Alliance. The strategy also outlines measures to support an AI-ready workforce and foster open-source AI solutions within the public sector.

This document presents a comprehensive survey of trustworthy physical intelligence, establishing a framework to address challenges such as physical safety, security, and integration across the system life cycle. It characterizes core capabilities, analyzes the role of physics in AI, introduces the Trustworthy Physical Intelligence Operationalization (T-PAIO) framework, and details the Trustworthy Physical Intelligence (T-PAI) governance framework.

This article proposes a Role-Evidence-Governance (REG) Framework to examine the use of social humanoid autonomous robots (SHARs) in K-12 schools. Through an integrative synthesis spanning educational robotics, learning theory, and AI governance, the framework evaluates the roles assigned to SHARs, the evidentiary warrants for those functions, and the governance conditions required for educational defensibility. It distinguishes between directly supported functions, adjacent-technology functions, and speculative high-risk applications while identifying nondelegable human responsibilities.

President Donald J. Trump signed an Executive Order replacing the term 'Artificial Intelligence' with 'Super Intelligence' across the executive branch to reflect technological advancements. The order directs departments and agencies to update official communications and requires the Assistant to the President for Science and Technology to propose a Federal definition for 'Super Intelligence'.

This document reviews Cheng Lin’s work on RoboHumanities, examining the intersections of AI, robotics, narratives, ethics, and philosophy through a collection of thirteen specialized essays. It focuses on the pedagogical, philosophical, and ontological shifts brought about by generative artificial intelligence and machine learning in human environments. The papers explore the history of educational automation, the cognitive role of digital interfaces, and critical competence regarding automated systems.

This paper presents a software-lifecycle framework for long-lived robot systems that decouples a fixed decision layer from versioned embodied capability modules and a runtime governance layer. Evaluating the system on six robosuite tasks, the authors report three negative results—including instability in peak task success across random seeds, performance limitations of time-sliced multi-policy rotation, and paralysis caused by an overly aggressive action-level safety shield—alongside measurements of promotion gates and rollback machinery.

This policy note examines the emerging field of biohybrid robotics and argues for an anticipatory regulatory framework to maximise positive societal impacts while mitigating potential risks. It reviews the currently fragmented and non-specific regulatory landscape, forecasts potential positive and negative societal implications including dual-use risks, and summarises findings from a stakeholder policy workshop on practical regulation and public trust.

This article proposes a conceptual framework called residual governance for responsible embodied artificial intelligence, addressing how systems act under inevitable computational, physical, and human-state approximations. It introduces two linked mechanisms—a residual ledger to record action-relevant uncertainties and an authority ledger to track who converts residual evidence into intervention. The paper argues that cognitive sovereignty is preserved not by eliminating algorithmic influence, but by making authority transitions accountable, contestable, and reversible.

This review analyzes the structural liability-coverage gap in co-located robotic surgery and telesurgery, arguing that the existing patchwork of separate insurance products leaves the field simultaneously over-insured and under-insured. It highlights unique legal, tort, and cyber-physical exposures arising from distributed actors across a surgical event, such as cross-jurisdictional licensing and connectivity liability. The authors contend that an integrated, single-contract liability insurance product with a single coverage trigger is needed to resolve these inter-insurer disputes and duplication costs.

The European AI Office is established within the European Commission to support trustworthy AI development and enforcement of the EU Artificial Intelligence Act. It oversees general-purpose AI (GPAI) models, coordinates member state authorities, and fosters an innovative and trustworthy AI ecosystem across Europe. The office works alongside advisory bodies like the European Artificial Intelligence Board, scientific panel, and stakeholder forum to implement regulatory and policy initiatives.

This academic opinion piece examines how anthropomorphic design in human-robot interaction influences human expectations and ethical relationships. Drawing on embodied cognition and Emmanuel Levinas's philosophy, the paper argues that designing social robots with strong human-like cues optimizes for functional social responsiveness while risking the totalization of alterity. Consequently, repeated interactions with predictable social agents may subtly constrain individuals' openness to the irreducible unpredictability of human others.

This paper introduces Queer-Inclusive Responsible Robotics (QIRR), an ethical framework organized around accountability and governance, privacy and safety, and representation and inclusion. Informed by queer ethics, intersectionality, and inclusive design, QIRR provides recommendations for inclusive interaction, identity-sensitive privacy, and governance across the robotics lifecycle. It positions queer inclusion as essential to responsible robotics and as a lens for addressing broader ethical, legal, and social implications.

This research paper examines the interplay between the physical capabilities of AI-driven smart robots, such as computer vision and heavy actuators, and the multi-layered safety and ethical guardrails that prevent autonomous harm. It outlines a hierarchical safety architecture ranging from physical hardware control up to government laws and legal liability. Additionally, the paper analyzes common public misconceptions regarding robot aggression, attributing them to teleoperation, kinematic recalibration, or manipulated media rather than true autonomous malice.

This study critically examines the rise of artificial intimacy and companion robots through the lens of Islamic legal philosophy (uṣūl al-fiqh). It argues that companion robots function as a form of technological substitution ('Digital Badal') that erodes the moral labor, mutual responsibility, and spiritual purposes of Islamic marriage (nikāḥ). Rejecting the argument that such devices can be permitted under necessity as a lesser harm, the paper proposes a comprehensive preventive policy grounded in blocking the means to harm (sadd al-dharāʾiʿ) alongside educational and community support measures.

This empirical study investigates Sparrow's ethical asymmetry hypothesis in human-robot interaction by testing whether humans are more inclined to condemn negative actions toward robots than to praise positive ones. Using an online questionnaire with manipulated moral permissibility and four cardinal virtue types, the research tests observer evaluations of human actors. The findings reveal a symmetrical, cubic relationship between perceived moral permissibility and perceived virtue scores, contradicting Sparrow's asymmetry hypothesis.

This document is a public consultation paper issued by the Ministry of Law and the Intellectual Property Office of Singapore seeking stakeholder views on the impact of artificial intelligence on Singapore's intellectual property regime. It covers key issues across copyright and patent law, including the scope of computational data analysis exceptions, copyright risk management for AI output, the nature of human creativity in AI-assisted works, and inventorship standards for AI-assisted inventions. The consultation runs from 26 August to 22 October 2026.

This paper critically examines six social robotics projects funded under the European Union's Horizon 2020 initiative that use AI-driven semi-autonomous robots for autism interventions. Mobilizing perspectives from critical disability studies, critical autism studies, and the autism community, the article argues that these projects apply neuronormative behavioral interventions to a neurominority group, risking epistemic injustice and harm. It ultimately questions whether ethical assistive technologies for autistic individuals can exist under these prevailing paradigms.

This document outlines the establishment, terms of reference, and submission guidelines for the Parliament of Australia's Joint Select Committee on Artificial Intelligence. Appointed on 20 August 2026, the committee is tasked with inquiring into and reporting on various aspects of AI, including economic opportunities, workforce impacts, regulatory adequacy, risks, and national security. The document provides information on how the public and organizations can make submissions and details the committee's reporting deadline of 30 November 2026.

This operational ethics framework addresses human-humanoid robot interaction by focusing on input-side governance, specifically how an embodied humanoid robot should decline high-risk human requests. Grounded in five working hypotheses from neo-humanist theory, the framework derives engineering artifacts including an A/B/C/D consequence-grading model, a dual-source split-key black box, an adversarial test suite for sycophancy, and a subjectivity-contraction indicator set. It includes scenario-specific rules across home, eldercare, healthcare, industrial, and public-service settings alongside a proposed standardization path.

The Foundational Ethics of Non-Humanoid Human–Robot Interaction provides a theoretical framework for non-humanoid robots such as industrial manipulators, AGVs, drones, and robot swarms operating in shared human spaces. Developed as Version 1.0 from He Yi's New Humanism theory lineage, it addresses three core differences from humanoid scenarios: the principal's absence, motion-based interaction channels, and multi-unit swarm objects. The framework establishes a seven-layer structure containing sub-theories like Bystander Sovereignty and Behavioral Legibility with quantifiable mechanisms designed for inclusion in standards and contracts.

This paper presents a ROS2-based self-adaptation approach building upon the MAPE-K framework to handle multiple simultaneous failures, cascading failures, and multiple resolving strategies in complex robotic systems. The authors introduce a domain-specific language (DSL) that allows designers to specify failure patterns, assign criticality levels, and enumerate adaptation strategies. Through a live dependency graph and a probabilistic planning algorithm, the approach performs root-cause analysis and strategy ranking to prioritize minimal and effective system adaptations.

This academic paper introduces SAFE-CHEM, an uncertainty-aware imitation learning framework designed to improve the safety and reliability of autonomous robotic manipulation in chemistry laboratories. The approach uses an ensemble of recurrent neural network-based policies to quantify epistemic uncertainty online through action variance, employing kernel density estimation to trigger a hybrid control switch to a rule-based backup controller when uncertainty exceeds a calibrated safety threshold. Evaluated across three fundamental laboratory manipulation tasks and demonstrated via zero-shot sim-to-real transfer, the framework improves task success rates and reduces critical safety violations compared to single-policy baselines.

This academic capstone and Science, Technology, and Society (STS) research project examines the technical design of a four-degree-of-freedom handwriting robotic arm alongside an ethical analysis of manufacturing automation and employment. The technical portion details the fabrication of a low-backlash, closed-loop robotic arm capable of fine-motor handwriting tasks, while the STS portion investigates labor polarization, deskilling, and psychological strain on industrial workers. The study concludes that mitigating the negative impacts of automation requires organizations to integrate proactive upskilling programs during technology deployment.

This research preprint develops a national-level decision-support framework and agent-based policy model for managing the long-run transition to an economy shaped by artificial intelligence and general-purpose robotic automation. It combines a mathematical system with agent-based simulations to evaluate policy packages involving automation rent taxes, statutory public capital ownership, social dividends, and staged work-time reduction. The study provides a calibration protocol, a roadmap with monitoring indicators, and a replication package to address deep uncertainty in labor markets and inequality.

This methodological study develops a monotone, redundancy-aware decision-burden method to provide explainable human oversight recommendations in human-robot collaboration. By replacing an unstable activity-normalized ratio with a pairwise overlap correction, the proposed Artificial Age Score (AAS) extension guarantees mathematical continuity and coordinatewise monotonicity. Through a controlled computational evaluation of 65,536 Sobol profiles, the corrected score achieved a held-out ROC AUC of 0.892, outperforming unadjusted logarithmic and linear baseline aggregation methods without triggering score decreases under input perturbation.

This systematic review examines how mobile service robots (MSRs) and social robots are defined in academic literature and explores their alignment with international standards, particularly ISO 13482:2014 and ISO/DIS 13482:2024. The authors analyze N = 51 articles to identify discrepancies between regulatory definitions and practical applications, noting that current standards often overlook vulnerable populations and non-physical risks. Ultimately, the paper proposes a refined MSR definition incorporating diverse embodiments, interaction modes, and user considerations to improve legal compliance and safety.

This academic article examines the ethical boundaries of introducing social assistive robots into healthcare through the lens of Katie Eriksson's theory of caritative caring. The authors argue that while robots can support routine tasks, safety, and autonomy, they lack the moral agency and compassionate intentionality required for true caritative caring. Therefore, robots should complement rather than replace human caregivers and relational care.

Diese Leitlinien legen den Umfang der Transparenzpflichten für Anbieter und Betreiber von KI-Systemen gemäß Artikel 50 des KI-Gesetzes fest. Sie wurden von der Europäischen Kommission angenommen, um den zuständigen Behörden sowie Anbietern und Betreibern praktische Orientierung zu bieten. Ziel ist es, die Einhaltung der Transparenzpflichten in kohärenter, wirksamer, verhältnismäßiger und einheitlicher Weise sicherzustellen. Die entsprechenden Transparenzpflichten gelten ab dem 2. August 2026.

These guidelines outline the scope of transparency obligations for providers and deployers of AI systems under Article 50 of the Artificial Intelligence Act. Adopted by the European Commission, the document aims to ensure a consistent, effective, proportionate, and uniform application of transparency requirements across the European Union. These transparency obligations are applicable starting from August 2, 2026.

The NSW AI Operational Policy sets out all-of-government requirements for the safe, lawful, and ethical use of artificial intelligence across New South Wales Government agencies and employees. It defines clear governance and assurance accountabilities, adopts Australia's AI Ethics Principles, and establishes mandatory rules for acceptable AI use, risk assessment, and incident reporting. The policy also mandates the registration of all AI use cases in the AI Assessment Framework Platform to improve transparency and support safe AI adoption and innovation.

This research paper introduces FoMo-FD (Flow-Matching World Model for Failure Detection), a method that learns nominal short-horizon visual dynamics with an action-conditioned flow-matching world model to detect low-level execution failures in surgical robot imitation policies. By scoring the inverse-transport nonconformity of observed endpoint latents and calibrating detection thresholds on successful rollouts, the method achieves robust window-level failure detection across multiple tasks without requiring failure demonstrations. Evaluations on simulation and real-world da Vinci Research Kit (dVRK) manipulation tasks demonstrate high failure detection rates with low false alarm rates.

This determination states that all foreign-produced advanced robotic devices pose an unacceptable risk to United States national security and the safety of U.S. persons, recommending their inclusion on the FCC's Covered List. It outlines the specific supply chain and cybersecurity vulnerabilities associated with networked robots, such as data exfiltration, remote disruption, and pre-installed backdoors. Furthermore, the document establishes definitions for advanced robotic devices and provides an exemption process through conditional approvals or specific Department of War determinations.

These advisory guidelines clarify how the Personal Data Protection Act 2012 (PDPA) applies to the development, deployment, and post-deployment phases of generative artificial intelligence models and systems in Singapore. They outline expectations for relying on the publicly available exception for web-scraping, providing AI-specific notifications and obtaining consent, allocating responsibilities among model providers, system providers, and system deployers, and handling individual access and correction requests.

These advisory guidelines clarify how Singapore's Personal Data Protection Act (PDPA) applies to the development and deployment of generative AI models and systems. They cover exceptions for publicly available data, notification and consent requirements, data protection responsibilities across the AI lifecycle (for model providers, system providers, and system deployers), and the handling of individual access and correction requests.

This research paper presents a distributed governance framework for autonomous multi-robot systems based on Sociocracy 3.0 (S3) principles and a three-tier consent protocol integrated with ROS 2. Evaluated through controlled simulations and the DROID-100 dataset, the framework demonstrates significant reductions in coordination conflicts and improvements in debuggability while maintaining real-time performance up to scaled team thresholds. The authors propose this architecture as an ethical, consent-driven approach to multi-agent robotic coordination in human-shared environments.

The National AI Strategy 2032 of the Republic of Cyprus serves as a strategic blueprint and roadmap to position Cyprus as a trusted AI hub in the Eastern Mediterranean. It outlines eight integrated national strategic objectives spanning governance, data foundations, public sector transformation, talent development, sovereign infrastructure, responsible AI ethics, and targeted industry adoption. The document details implementation approaches, priority sectors for leadership, regulatory alignment with European frameworks such as the EU AI Act, and an innovation value chain to accelerate productivity and economic resilience.