Artificial intelligence: laws, standards and reference sources
The laws, international instruments, technical standards, open protocols and public data that frame artificial intelligence, each with its date, its status and its primary source.
This page lists the main texts and sources that frame artificial intelligence on 8 October 2026: the Artificial Intelligence Act of the European Union and its amended timeline, the international instruments, the rules of the United Arab Emirates and of the Asian jurisdictions where My NEO Group is present, the technical standards, the open protocols for AI agents and the public data sources. Each entry gives the official name, the date, the status and a link to the primary source.
2026 edition, reviewed 64 primary sources in 7 groups
Declaration of interest
My NEO Group develops NEO AI, an artificial intelligence system, through its company NEO Labs. This page is published by My NEO Group as a reference. It is not legal advice. It does not assess NEO AI, or any other product, against the texts listed.
What the law calls an AI system
The Artificial Intelligence Act of the European Union defines an AI system in Article 3(1): a machine-based system that operates with varying levels of autonomy, that may adapt after it is deployed, and that infers from the input it receives how to generate outputs such as predictions, content, recommendations or decisions.
The Recommendation of the OECD on Artificial Intelligence uses a close definition: a machine-based system that infers, from the input it receives, how to generate outputs that can influence physical or virtual environments. The OECD adds that AI systems vary in their autonomy and in their adaptiveness after deployment. The OECD updated this definition on 8 November 2023.
Article 3 of the AI ActOECD Recommendation on Artificial IntelligenceGlossary: names and terms, each with one definition
European Union
The Artificial Intelligence Act is the binding law of the European Union on AI. Its timeline was amended in July 2026.
| Text or source | Type | Date | Status |
|---|---|---|---|
| Artificial Intelligence Act - Regulation (EU) 2024/1689Regulation of 13 June 2024 laying down harmonised rules on artificial intelligence in the European Union. It was published in the Official Journal on 12 July 2024 and entered into force on 1 August 2024. It applies in stages, set out in the timeline below.Primary sourceEuropean Union# | Law | In force, applies in stages | |
| Digital Omnibus on AI - Regulation (EU) 2026/1744Regulation of 8 July 2026 that amends the Artificial Intelligence Act. It was published in the Official Journal on 24 July 2026 and entered into force on 27 July 2026. It moved the high-risk requirements to 2 December 2027 for the systems of Annex III and to 2 August 2028 for the products of Annex I.Primary sourceEuropean Union# | Law | In force | |
| Human oversight - Article 14 of the AI ActArticle 14 requires high-risk AI systems to be designed so that natural persons can effectively oversee them while they are in use. It concerns high-risk systems only, and it applies with the high-risk requirements, from 2 December 2027 or from 2 August 2028 depending on the system.Primary sourceEuropean Union# | Law | In force, applies in stages | |
| General-Purpose AI Code of PracticeVoluntary code published on 10 July 2025 to help providers of general-purpose AI models comply with the AI Act. It has three chapters: Transparency, Copyright, and Safety and Security. The European Commission and the AI Board confirmed it as an adequate voluntary tool on 1 August 2025.Primary sourceEuropean Union# | Guideline | Voluntary |
Timeline of the Artificial Intelligence Act, as amended
The table gives the dates of application of the Artificial Intelligence Act as published by the European Commission after the amendment of July 2026. Texts written before July 2026 give earlier dates for the high-risk requirements: they are out of date.
- The Artificial Intelligence Act enters into force.
- The prohibited AI practices and the AI literacy obligations apply.
- The governance rules and the obligations for general-purpose AI models apply.
- Most rules apply, including the transparency rules of Article 50. Enforcement by the AI Office and by national authorities starts.
- The prohibitions added by the Digital Omnibus on AI apply. Systems that generate synthetic content and were placed on the market before 2 August 2026 must comply with the marking rule of Article 50(2).
- Each Member State should have at least one AI regulatory sandbox in operation.
- The high-risk requirements apply to the AI systems of Annex III.
- The high-risk requirements apply to AI in the products covered by Annex I.
International instruments
Recommendations, treaties and processes agreed between states. Most of them are not binding.
| Text or source | Type | Date | Status |
|---|---|---|---|
| OECD Recommendation on Artificial Intelligence (OECD AI Principles)Adopted by the OECD Council on 22 May 2019 and revised on 8 November 2023 and on 3 May 2024. The OECD register of legal instruments lists 38 OECD Members, 12 non-Members and the European Union as adherents.Primary sourceOECD# | Recommendation | Voluntary | |
| UNESCO Recommendation on the Ethics of Artificial IntelligenceAdopted by the General Conference of UNESCO on 23 November 2021. UNESCO states that its 193 Member States adopted it. Member States apply it on a voluntary basis.Primary sourceUNESCO# | Recommendation | Voluntary | |
| Council of Europe Framework Convention on Artificial Intelligence and Human Rights, Democracy and the Rule of Law (CETS No. 225)Opened for signature in Vilnius on 5 September 2024. Its entry into force requires five ratifications, including three by member States of the Council of Europe. On the date of this edition, the chart of the Treaty Office showed one ratification, by the European Union on 15 May 2026.Primary sourceCouncil of Europe# | Treaty | Not in force | |
| Hiroshima Process International Guiding Principles and International Code of Conduct for Organizations Developing Advanced AI SystemsTwo voluntary instruments published by the G7 on 30 October 2023, under the Hiroshima AI Process started during the Japanese presidency of the G7.Primary sourceG7# | Guideline | Voluntary | |
| Hiroshima AI Process Reporting FrameworkVoluntary framework launched on 7 February 2025 and hosted by the OECD. Organisations use it to report on how they apply the Hiroshima Process International Code of Conduct.Primary sourceG7 and OECD# | Guideline | Voluntary | |
| Global Digital CompactAdopted in New York on 22 September 2024 with the Pact for the Future. It committed the Member States of the United Nations to set up an international scientific panel on AI and a global dialogue on AI governance.Primary sourceUnited Nations# | Recommendation | Voluntary | |
| Independent International Scientific Panel on Artificial IntelligenceEstablished by General Assembly resolution A/RES/79/325 of 26 August 2025. Its 40 members were appointed in February 2026 and serve in their personal capacity. The Panel released a preliminary report on 1 July 2026.Primary sourceUnited Nations# | Public body | Active | |
| Global Dialogue on AI GovernanceEstablished by the same General Assembly resolution of 26 August 2025. Its first session was held in Geneva on 6 and 7 July 2026. The second session is scheduled for 3 and 4 May 2027 in New York.Primary sourceUnited Nations# | Public body | Active | |
| AI Safety Summit 2023Held at Bletchley Park, in the United Kingdom, on 1 and 2 November 2023. The Bletchley Declaration was published on 1 November 2023.Primary sourceUnited Kingdom# | Summit | Held | |
| AI Seoul Summit 2024Held on 21 and 22 May 2024, co-hosted by the Republic of Korea and the United Kingdom. The Seoul Declaration was issued on 21 May 2024.Primary sourceRepublic of Korea# | Summit | Held | |
| AI Action SummitHeld in Paris on 10 and 11 February 2025. A statement on inclusive and sustainable artificial intelligence was published on 11 February 2025.Primary sourceFrance# | Summit | Held | |
| India AI Impact Summit 2026Held in New Delhi in February 2026. The declaration of the summit is dated 19 February 2026.Primary sourceIndia# | Summit | Held | |
| Geneva AI Summit 2027Switzerland announced on 19 February 2026 that it will host the next summit of the series. The Swiss Federal Office of Communications gives the dates 21 and 22 June 2027, in Geneva.Primary sourceSwitzerland# | Summit | Announced |
United Arab Emirates
The United Arab Emirates frames artificial intelligence through national strategies, a charter of principles, a federal authority and the rules of its emirates and financial centres. My NEO Group has its headquarters in Dubai.
| Text or source | Type | Date | Status |
|---|---|---|---|
| National Artificial Intelligence Strategy 2031Adopted by the Cabinet of the United Arab Emirates on 21 April 2019. It follows the UAE Strategy for Artificial Intelligence launched in October 2017. Its stated aim is to make the country a world leader in artificial intelligence by 2031.Primary sourceUnited Arab Emirates# | Strategy | Policy document | |
| Artificial Intelligence and Data AuthorityFederal authority whose establishment was approved on 14 June 2026. It reports to the Cabinet and brings together the functions of the Office of Artificial Intelligence, Digital Economy and Remote Work Applications, of the digital government sector of the telecommunications regulator and of the UAE Data Office. Its mandate includes proposing policies and legislation on AI.Primary sourceUnited Arab Emirates# | Public body | Active | |
| The UAE Charter for the Development and Use of Artificial IntelligenceCharter of twelve principles published in 2024. The principles include safety, data privacy, transparency, human oversight, and governance and accountability. It is not a law.Primary sourceUnited Arab Emirates# | Charter | Voluntary | |
| Position of the United Arab Emirates on artificial intelligence policy internationallyPolicy position approved by the Cabinet and published in October 2024. It rests on six principles: advancement, cooperation, community, ethics, sustainability and security.Primary sourceUnited Arab Emirates# | Strategy | Policy document | |
| DIFC Data Protection Regulations - Regulation 10, Personal Data Processed Through Autonomous and Semi-Autonomous SystemsRegulation of the Dubai International Financial Centre, in force since 1 September 2023. It sets rules for the processing of personal data by autonomous and semi-autonomous systems and defines the roles of deployer and operator of such a system. It applies within the DIFC.Primary sourceDubai International Financial Centre# | Regulation | In force | |
| Dubai Universal Blueprint for Artificial IntelligencePlan of the Government of Dubai launched on 29 April 2024 to speed up the adoption of AI. Its first phase includes the appointment of a Chief Artificial Intelligence Officer in each government entity of Dubai.Primary sourceDubai# | Strategy | Policy document | |
| Dubai AI SealCertification launched on 20 January 2025 by the Dubai Centre for Artificial Intelligence for technology companies licensed in Dubai that provide AI products or services.Primary sourceDubai# | Guideline | Voluntary | |
| AI Policy for Government Entities of DubaiPolicy announced on 22 April 2025, during Dubai AI Week, by Digital Dubai and the Dubai Future Foundation. It sets one framework for the use of AI across the government departments of Dubai.Primary sourceDubai# | Guideline | Policy document | |
| Artificial Intelligence and Advanced Technology Council (AIATC)Council of the Emirate of Abu Dhabi, established by a law announced on 22 January 2024. It develops and implements the policies and strategies of Abu Dhabi on research, infrastructure and investment in AI and advanced technology.Primary sourceAbu Dhabi# | Public body | Active | |
| Abu Dhabi Government Digital Strategy 2025-2027Strategy announced on 21 January 2025, with AED 13 billion planned over 2025-2027. Its stated goal is a government that is AI-native across all digital services by 2027.Primary sourceAbu Dhabi# | Strategy | Policy document |
On the official pages read for this edition, no federal law dedicated to artificial intelligence was found for the United Arab Emirates.
Asia
Laws, measures and guidelines of the Asian jurisdictions where My NEO Group is present, and of Singapore, whose frameworks are a reference in the region. For texts with no official English title, the English name is a translation.
| Text or source | Type | Date | Status |
|---|---|---|---|
| Interim Measures for the Management of Generative Artificial Intelligence ServicesMeasures issued jointly by seven authorities led by the Cyberspace Administration of China, promulgated on 10 July 2023 and in force since 15 August 2023.Primary sourceChina# | Regulation | In force | |
| Measures for Labeling AI-Generated and Synthetic ContentMeasures issued by four authorities, published on 14 March 2025 and in force since 1 September 2025. They set rules for the labeling of content generated or synthesised by AI.Primary sourceChina# | Regulation | In force | |
| Amendment of the Cybersecurity Law of the People's Republic of China (new Article 20 on artificial intelligence)Decision adopted on 28 October 2025 by the Standing Committee of the National People's Congress and in force since 1 January 2026. It adds to the Cybersecurity Law an article on artificial intelligence. It is an amendment, not a separate law on AI.Primary sourceChina# | Law | In force | |
| Hong Kong Generative Artificial Intelligence Technical and Application GuidelineGuideline promulgated on 15 April 2025 by the Digital Policy Office of the Government of Hong Kong. It gives operational guidance to developers, service providers and users. The version online on the date of this edition is version 1.1, dated December 2025.Primary sourceHong Kong# | Guideline | Voluntary | |
| Artificial Intelligence: Model Personal Data Protection FrameworkFramework published on 11 June 2024 by the Privacy Commissioner for Personal Data of Hong Kong, for organisations that procure, implement and use AI systems. One of its four areas is risk assessment and human oversight.Primary sourceHong Kong# | Guideline | Voluntary | |
| Consumer Protection in respect of Use of Generative Artificial IntelligenceCircular issued on 19 August 2024 by the Hong Kong Monetary Authority to all authorized institutions.Primary sourceHong Kong# | Guideline | Supervisory guidance | |
| Act on Promotion of Research and Development, and Utilization of Artificial Intelligence-related Technology (Act No. 53 of 2025)Act of 4 June 2025, fully in force since 1 September 2025. It provides for an Artificial Intelligence Strategic Headquarters under the Cabinet and for a basic plan on artificial intelligence.Primary sourceJapan# | Law | In force | |
| Japan's Second Artificial Intelligence Basic PlanPlan adopted by the Cabinet of Japan on 14 July 2026. It replaced the first plan, adopted on 23 December 2025.Primary sourceJapan# | Strategy | Policy document | |
| Framework Act on the Development of Artificial Intelligence and the Creation of a Foundation for Trust (Act No. 20676)Framework law of the Republic of Korea on artificial intelligence, promulgated on 21 January 2025 and amended by an act promulgated on 20 January 2026. The English title is the reference translation of the Korea Legislation Research Institute.Primary sourceRepublic of Korea# | Law | Published | |
| India AI Governance GuidelinesGuidelines unveiled on 5 November 2025 by the Ministry of Electronics and Information Technology of India, under the IndiaAI Mission. They set out seven guiding principles and recommendations across six pillars of AI governance.Primary sourceIndia# | Guideline | Voluntary | |
| IndiaAI MissionNational programme approved by the Union Cabinet of India on 7 March 2024. It has seven components, among them compute capacity, datasets, skills, start-up financing and safe and trusted AI.Primary sourceIndia# | Strategy | Policy document | |
| Law on Artificial Intelligence of VietnamLaw passed by the National Assembly of Vietnam on 10 December 2025, with 8 chapters and 35 articles. It took effect on 1 March 2026.Primary sourceVietnam# | Law | In force | |
| Law on Digital Technology Industry of VietnamLaw passed by the National Assembly of Vietnam on 14 June 2025, with effect from 1 January 2026. It requires human oversight of AI systems and classifies them by risk. It also introduces a legal framework for digital assets.Primary sourceVietnam# | Law | In force | |
| Draft principles of a law on artificial intelligence of ThailandDraft principles published for public consultation on 11 June 2025 by the Electronic Transactions Development Agency of Thailand. They focus on the oversight of high-risk uses of AI. On the official pages read for this edition, Thailand had not enacted a law on AI.Primary sourceThailand# | Law | Draft | |
| Model AI Governance Framework for Generative AIFramework released on 30 May 2024 by the AI Verify Foundation and the Infocomm Media Development Authority of Singapore. It sets out nine dimensions, among them accountability, testing and assurance, security and content provenance.Primary sourceSingapore# | Guideline | Voluntary | |
| Model AI Governance Framework for Agentic AIFramework launched on 22 January 2026 by the Infocomm Media Development Authority of Singapore. It gives guidance in four areas: assess and bound risks early, keep humans accountable, apply technical controls across the life of an agent, and support the responsibility of end users.Primary sourceSingapore# | Guideline | Voluntary |
Technical standards and risk frameworks
Standards and frameworks used to build, manage and assess AI systems. ISO, IEC and IEEE standards are published documents under copyright; this page gives their subject only.
| Text or source | Type | Date | Status |
|---|---|---|---|
| ISO/IEC 22989:2022 - Artificial intelligence concepts and terminologyStandard published in July 2022. It establishes the terminology of artificial intelligence and describes its concepts. Two amendments were in preparation on the date of this edition.Primary sourceISO and IEC# | Standard | Current edition | |
| ISO/IEC 42001:2023 - Artificial intelligence management systemStandard published in December 2023. It specifies the requirements for establishing, implementing, maintaining and improving an AI management system in an organisation.Primary sourceISO and IEC# | Standard | Current edition | |
| ISO/IEC 23894:2023 - Guidance on risk management for artificial intelligenceStandard published in February 2023. It gives guidance on how organisations that develop, deploy or use AI can manage the risks specific to AI.Primary sourceISO and IEC# | Standard | Current edition | |
| ISO/IEC 42005:2025 - AI system impact assessmentStandard published in May 2025. It gives guidance to organisations that assess how an AI system may affect individuals, groups or society.Primary sourceISO and IEC# | Standard | Current edition | |
| ISO/IEC 42006:2025 - Requirements for bodies that audit and certify AI management systemsStandard published in July 2025. It sets requirements for the bodies that audit and certify AI management systems under ISO/IEC 42001.Primary sourceISO and IEC# | Standard | Current edition | |
| NIST AI Risk Management Framework (AI RMF 1.0)Voluntary framework released on 26 January 2023 by the National Institute of Standards and Technology of the United States, to help organisations manage the risks of AI products, services and systems. NIST states that the framework is being revised.Primary sourceUnited States# | Guideline | Voluntary | |
| NIST AI 600-1 - Generative Artificial Intelligence ProfileProfile of the AI Risk Management Framework for generative AI, published on 26 July 2024.Primary sourceUnited States# | Guideline | Voluntary | |
| IEEE 7001-2021 - Standard for Transparency of Autonomous SystemsStandard published on 4 March 2022. It defines measurable levels of transparency for autonomous systems, so that a system can be assessed.Primary sourceIEEE# | Standard | Current edition | |
| OWASP GenAI LLM Top 10 2026Community list of the most critical security risks of applications built on large language models, in its edition dated 3 August 2026. The previous edition, the OWASP Top 10 for LLM Applications 2025, is dated 17 November 2024.Primary sourceOWASP# | Guideline | Current edition | |
| OWASP Top 10 for Agentic Applications for 2026Community list of ten security risks specific to applications built on AI agents, published on 9 December 2025. The entries include agent goal hijack, tool misuse, and identity and privilege abuse.Primary sourceOWASP# | Guideline | Current edition |
Open protocols for AI agents
Open specifications that let AI applications and AI agents connect to tools, to data and to one another.
| Text or source | Type | Date | Status |
|---|---|---|---|
| Model Context Protocol (MCP)Open protocol that connects applications built on large language models to external data sources and tools. Anthropic announced it on 25 November 2024 and donated it on 9 December 2025 to the Agentic AI Foundation, under the Linux Foundation. The current revision of the specification is dated 28 July 2026.Primary sourceLinux Foundation# | Open protocol | Current edition | |
| Agent2Agent (A2A) ProtocolOpen protocol for communication between AI agents built by different parties. Google announced it on 9 April 2025, and the Linux Foundation launched the Agent2Agent project on 23 June 2025. Version 1.0 was announced on 12 March 2026.Primary sourceLinux Foundation# | Open protocol | Current edition | |
| Agentic AI Foundation (AAIF)Foundation formed on 9 December 2025 under the Linux Foundation to host open projects for AI agents. Its three founding projects are the Model Context Protocol, goose and AGENTS.md.Primary sourceLinux Foundation# | Public body | Active |
Public data and reference reports
Sources of data on artificial intelligence that publish their method. Figures are those of the publisher named, for the period named.
| Text or source | Type | Date | Status |
|---|---|---|---|
| International AI Safety Report 2026Second edition of the report, published on 3 February 2026. More than 100 experts contributed to it, under an advisory panel with nominees from over 30 countries and international organisations. The first edition was published on 29 January 2025.Primary sourceInternational# | Report | Current edition | |
| The 2026 AI Index ReportAnnual report of the Stanford Institute for Human-Centered Artificial Intelligence, in its edition of April 2026. Its nine chapters cover research, technical performance, responsible AI, the economy, science, medicine, education, policy and public opinion.Primary sourceStanford University# | Report | Current edition | |
| Epoch AI - data on AIPublic databases of a research nonprofit on AI models, AI hardware, computing clusters and benchmarks. Epoch AI states that its data may be reused with attribution.Primary sourceEpoch AI# | Data source | Live | |
| OECD.AI Policy ObservatoryPlatform of the OECD that gives access to a database of national AI policies and initiatives, updated by official contact points of countries and international organisations.Primary sourceOECD# | Data source | Live | |
| AI Incidents and Hazards Monitor (AIM)Monitor of the OECD that documents AI incidents and hazards reported in the news media. The OECD labels it as a beta version.Primary sourceOECD# | Data source | Live | |
| AI Incident DatabaseDatabase of the Responsible AI Collaborative that indexes harms and near harms caused by deployed AI systems. Incidents are submitted by the public and reviewed by editors.Primary sourceResponsible AI Collaborative# | Data source | Live | |
| Key Questions on Energy and AIReport of the International Energy Agency published on 16 April 2026. The IEA estimates the electricity consumption of data centres at 485 TWh in 2025 and projects about 950 TWh in 2030. It follows the report Energy and AI of 10 April 2025.Primary sourceInternational Energy Agency# | Report | Current edition | |
| Task-completion time horizons of frontier AI modelsMeasure published by the research nonprofit METR: the length of the tasks, counted in the time a human expert needs, that an AI agent completes with a given reliability. Its study of 19 March 2025 found that this length had doubled about every seven months over six years. METR updated its method on 29 January 2026.Primary sourceMETR# | Data source | Live |
Dates after this edition
Dates announced by the sources above that fall after the date of this edition. They are the dates published on 8 October 2026 and may change.
- The prohibitions added by the Digital Omnibus on AI apply. Systems that generate synthetic content and were placed on the market before 2 August 2026 must comply with the marking rule of Article 50(2).Details
- Second session of the Global Dialogue on AI Governance of the United Nations, in New York, on 3 and 4 May 2027.Details
- Geneva AI Summit 2027, on 21 and 22 June 2027.Details
- Each Member State should have at least one AI regulatory sandbox in operation.Details
- The high-risk requirements apply to the AI systems of Annex III.Details
- The high-risk requirements apply to AI in the products covered by Annex I.Details
My NEO Group and artificial intelligence
My NEO Group works in artificial intelligence through NEO Labs, which develops NEO AI. NEO AI is described on its own page of this site: an Intelligence Operating System that coordinates AI models and AI agents, in which some actions always require a human approval.
This page lists public texts and sources. It makes no statement on how NEO AI, or any product, relates to them.
NEO AI: definition, components and statusNEO Labs: the company behind NEO AIGlossary: names and terms, each with one definition
How this page was made
- Each entry was read on the official page linked, on 8 October 2026. A fact that could only be confirmed on a secondary source is not printed.
- Each entry gives the official English name when one exists. For texts with no official English title, the English name is a translation.
- The status is the one shown by the primary source on the date of this edition. Laws and timelines change: check the primary source before relying on a date.
- Texts are summarised in the words of this site. The summaries do not replace the texts.
- This page is information. It is not legal advice.
Rules of this sectionFacts as JSONThis page as Markdown
Regulated services are provided only by licensed entities, in the jurisdictions where they are authorised. Nothing on this site is an offer of securities or investment advice.