Standards & Ethics
AI Act support instruments, ISO and European AI standards, and international ethical frameworks.
Standards, voluntary guidance and standards-development programmes are distinct. Inclusion here does not establish harmonised status under the EU AI Act. Summaries remain draft; official catalogue pages may charge for the full standard.
Showing 24 of 24 standards, frameworks and support instruments
AI Act Compliance Checker / Explorer (opens in a new tab)
The official platform provides an AI Act Explorer for chapters, annexes and recitals and a Compliance Checker to help evaluate requirements for AI systems and GPAI models. Tool results do not constitute certification or a binding legal determination.
AI Act Service Desk and Single Information Platform (opens in a new tab)
Official Commission information hub offering AI Act guidance, interactive tools and a Service Desk for questions. An information service, not a law or certification.
Voluntary initiative supporting preparation for the AI Act through community exchange and company pledges. It is not a binding legal instrument.
Assessment List for Trustworthy Artificial Intelligence (ALTAI) (opens in a new tab)
Voluntary self-assessment checklist for developers and deployers, based on the seven requirements of the European trustworthy-AI guidelines. It is not certification or a technical standard.
BS EN ISO/IEC 42001:2026 — European AI management-system standard (opens in a new tab)
European and UK adoption of ISO/IEC 42001:2023, published by BSI on 31 March 2026. BSI states that requirements and guidance are technically unchanged. Provides an AI management-system framework; European adoption alone does not establish harmonised EU AI Act status.
CEN-CENELEC JTC 21 — European AI standardisation (opens in a new tab)
European standards-development programme supporting AI governance and EU legislation. JTC 21 is a technical committee, not an adopted standard; this page does not establish that a particular standard is harmonised or cited in the EU Official Journal.
Code of Practice on Transparency of AI-Generated Content (opens in a new tab)
Voluntary code offering practical measures for marking, detection and labelling of AI-generated content under Article 50. The Commission page identifies the final code and confirmation of its adequacy.
Commission guidelines on Article 50 transparency (opens in a new tab)
Guidelines on transparency obligations for providers and deployers of certain AI systems, explaining Article 50 duties. Distinct from the voluntary code on AI-generated content.
Commission guidelines on GPAI obligations (opens in a new tab)
Guidelines for providers of general-purpose AI models explaining which providers are in scope and how their AI Act obligations apply. Interpretive guidance, not a separate law.
Commission guidelines on high-risk classification — draft (opens in a new tab)
Draft Commission guidelines on the classification of high-risk AI systems under Article 6, including practical examples. The linked institutional page describes a draft consultation document; this entry does not claim a final version has been adopted.
Commission guidelines on prohibited AI practices (opens in a new tab)
Non-binding Commission guidance interpreting the prohibited AI practices in Article 5, with legal explanations and practical examples.
EN 18286:2026 — Quality management system for EU AI Act regulatory purposes (opens in a new tab)
European standard translating Article 17 quality-management requirements into practical, auditable requirements for providers of high-risk AI systems. CEN-CENELEC reports its citation in the EU Official Journal as a harmonised standard in October 2026. Presumption of conformity concerns the corresponding provisions, not blanket compliance with the entire AI Act; the applicable Official Journal citation and scope require legal review.
Ethics Guidelines for Trustworthy AI (opens in a new tab)
European Commission AI High-Level Expert Group guidance on lawful, ethical and robust AI. Seven requirements cover oversight, robustness, privacy, transparency, fairness, societal well-being and accountability. Voluntary expert guidance, not a technical standard.
GPAI Code of Practice (opens in a new tab)
Final voluntary General-Purpose AI Code of Practice supporting model providers with AI Act transparency, copyright and safety/security obligations. The code does not replace the Act.
HUDERIA — Risk and impact assessment of AI systems (opens in a new tab)
Council of Europe non-binding guidance for assessing AI risks and impacts on human rights, democracy and the rule of law. Includes a methodology and a context-based risk analysis approach.
ISO/IEC 22989:2022 — AI concepts and terminology (opens in a new tab)
Establishes terminology and concepts for artificial intelligence, providing a shared vocabulary for technical and governance discussions.
ISO/IEC 23894:2023 — Guidance on AI risk management (opens in a new tab)
Guidance for managing risks associated with AI and integrating risk management into AI-related activities.
ISO/IEC 38507:2022 — Governance implications of AI (opens in a new tab)
Guidance for governing bodies on the implications of using AI within organisations, including oversight and responsible decision-making.
ISO/IEC 42001:2023 — AI management systems (opens in a new tab)
Requirements for establishing, implementing, maintaining and continually improving an AI management system within an organisation. An international standard, not by itself proof of EU AI Act compliance.
ISO/IEC TR 24027:2021 — Bias in AI systems and AI-aided decision-making (opens in a new tab)
Technical report addressing bias in AI systems, including approaches to measurement and assessment across the AI life cycle.
ISO/IEC TR 24028:2020 — Overview of trustworthiness in AI (opens in a new tab)
Technical report surveying trustworthiness in AI, including transparency, reliability, safety, security and privacy. Consult the catalogue for revisions and current edition status.
NIST AI Risk Management Framework (AI RMF 1.0) (opens in a new tab)
Voluntary US framework, NIST AI 100-1, for incorporating trustworthiness and managing risks throughout the AI lifecycle. Organised around Govern, Map, Measure and Manage. It is not an EU harmonised standard or proof of EU AI Act compliance.
OECD AI Principles — Recommendation of the Council on Artificial Intelligence (opens in a new tab)
Intergovernmental recommendation on trustworthy AI that respects human rights and democratic values. Adopted in 2019 and amended in 2024; includes values-based principles and policy recommendations.
UNESCO Recommendation on the Ethics of Artificial Intelligence (opens in a new tab)
International ethical recommendation adopted in 2021, addressing human dignity and rights, fairness, gender equality, inclusion, well-being and environmental protection. Non-binding international guidance.