AI Center › Global AI Governance Register
1ArtificialIntelligence · The Global AI Ecosystem — living dataset
Global AI Governance Register
The Global AI Governance Register answers one question for each country or bloc: what law or rule governs artificial intelligence there, who enforces it, whether it binds you yet, and where to read the official text. This release covers 15 jurisdictions, plus 2 international agreements that countries adopt. Every cell names the authority it came from and the date it was checked.
It is a neutral directory, not legal advice. Every entry was re-checked against its official source before publication, and anything that could not be confirmed against a primary source was left out rather than published unverified. The order is alphabetical within each group and implies no ranking.
Informational only. Not legal advice. Each entry states what the official text provides — read that text before relying on it. Laws and their status change; every row shows its own last-checked date.
Listed alphabetically within each group. The order implies no ranking — no country's approach is presented as better or worse than another's. · 17 jurisdictions and international agreements in this release.
| Country or bloc | Region | How it regulates AI | Who enforces it | Does it bind yet? |
|---|---|---|---|---|
| Australia | Oceania | Voluntary guidance | Office of AI, Department of the Prime Minister and Cabinet, established 2026-07-15, coordinating across government to design and legislate an Australian AI standard including mandatory requirements for large AI data centres; National AI Centre retains the guidance function. | No — voluntary |
| Brazil | South America | One broad AI law | Autoridade Nacional de Proteção de Dados (ANPD), proposed to coordinate the Sistema Nacional de Regulação e Governança de IA (SIA); bill under review in the Câmara dos Deputados | Not yet — still a draft |
| Canada | Northern America | Voluntary guidance | Innovation, Science and Economic Development Canada (ISED); Minister of Artificial Intelligence and Digital Innovation | No — the bill lapsed |
| Chile | South America | One broad AI law | Ministerio de Ciencia, Tecnología, Conocimiento e Innovación (MinCiencia); bill under review in the Senado (Comisión Desafíos del Futuro, Ciencia, Tecnología e Innovación); proposed oversight by the Agencia de Protección de Datos Personales, advised by a Consejo Asesor Técnico de Inteligencia Artificial | Not yet — still a draft |
| China | Eastern Asia | Through existing sector rules | For the 2026 anthropomorphic-interaction measures: CAC + National Development and Reform Commission + Ministry of Industry and Information Technology + Ministry of Public Security + State Administration for Market Regulation (SAMR replaces the education/science/broadcast ministries that co-issued the 2023 measures). | Yes — in force |
| European Union | Europe | One broad AI law | National competent authorities designated by each Member State (Art. 70), the European AI Office (Art. 64) and the European AI Board (Art. 65); the Commission has exclusive powers over general-purpose AI models under Chapter V (Art. 88(1)). From 27 July 2026, Arts. 75a–75d inserted by Regulation (EU) 2026/1744 give the AI Office direct supervisory and enforcement powers, including fines applying Art. 99(3)–(7), over certain AI systems within its competence. | Partly — phasing in |
| Israel | Western Asia | Voluntary guidance | Ministry of Innovation, Science and Technology with the Ministry of Justice (Office of Legal Counsel and Legislative Affairs, Economic Law Department); implementation via sectoral regulators | No — voluntary |
| Japan | Eastern Asia | One broad AI law | Cabinet — Artificial Intelligence Strategy Headquarters (chaired by the Prime Minister); Minister for AI Strategy | Yes — in force |
| Nigeria | Sub-Saharan Africa | National strategy only | National Information Technology Development Agency (NITDA) / National Centre for Artificial Intelligence and Robotics (NCAIR), under the Federal Ministry of Communications, Innovation and Digital Economy (FMCIDE) | No — voluntary |
| Saudi Arabia | Western Asia | Voluntary guidance | Saudi Data and Artificial Intelligence Authority (SDAIA) | No — voluntary |
| Singapore | South-Eastern Asia | Voluntary guidance | Infocomm Media Development Authority (IMDA) and Personal Data Protection Commission (PDPC); AI Verify Foundation | non-binding at national level; sector regulators impose binding obligations that reach AI, notably Monetary Authority of Singapore requirements for financial institutions |
| South Korea (Republic of Korea) | Eastern Asia | One broad AI law | Ministry of Science and ICT (MSIT) | Yes — in force |
| United Kingdom | Europe | Through existing sector rules | Department for Science, Innovation and Technology (DSIT); existing sector regulators (ICO, Ofcom, FCA, CMA, MHRA, etc.); AI Security Institute (formerly AI Safety Institute, renamed 14 Feb 2025) | No — voluntary |
| United States — Colorado | Northern America | One broad AI law | Colorado Attorney General (Department of Law) | Adopted, but not applying yet |
| United States — federal | Northern America | Through existing sector rules | Sectoral federal agencies (FTC, EEOC, CFPB, FDA) applying existing law, coordinated with NIST and the OMB / White House Office of Science and Technology Policy | non-binding on private developers; no federal AI statute has been enacted, and the obligations that exist are directed at federal agencies through executive orders |
| Council of Europe | International | Binding treaty | Council of Europe (Secretary General as depositary; Treaty Office); interim custodian Committee CDNET (from 1 Jan 2026); Conference of the Parties upon entry into force | Adopted, but not applying yet |
| OECD | International | Voluntary guidance | Organisation for Economic Co-operation and Development (OECD); OECD.AI Policy Observatory; Committee on Digital Economy Policy | in force as a Recommendation, and not legally binding on Adherents — the status the OECD itself records for the instrument |
Open data
All 17 rows are published as an openly licensed dataset — free to download, reuse and republish with attribution under CC BY 4.0 ↗.
Attribute as: 1BusinessWorld AI Center, "Global AI Governance Register", https://1businessworld.com/ai-center/global-ai-governance-register/ (version as of July 28, 2026), licensed CC BY 4.0.
Jurisdictions — 15 places whose rules govern AI
A jurisdiction is a place whose law can bind you: a country, a state, or a union with legal authority over a territory. Each row records what governs AI there, who enforces it, and whether it binds yet.
Australia (AU)
Oceania
- The law or rule
- Guidance for AI Adoption (National AI Centre, released October 2025, six essential practices), the current primary government guidance; the Voluntary AI Safety Standard and its 10 guardrails remain available as a more detailed control catalogue.
- What kind of rule
- voluntary standard / guidance
- Official reference
- Voluntary AI Safety Standard (DISR, published 5 Sep 2024, 10 guardrails); Guidance for AI Adoption (National AI Centre, 21 Oct 2025, six essential practices — consolidates/supersedes the 10 guardrails); National AI Plan (Australian Government, 2 Dec 2025)
- Does it bind yet?
- No — voluntaryStated in Voluntary AI Safety Standard (DISR, published 5 Sep 2024, 10 guardrails); Guidance for AI Adoption (National AI Centre, 21 Oct 2025, six essential practices — Austrade (Australian Trade and Investment Commission), Australian Government — reporting the DISR National AI Plan ↗
- Who enforces it
- Office of AI, Department of the Prime Minister and Cabinet, established 2026-07-15, coordinating across government to design and legislate an Australian AI standard including mandatory requirements for large AI data centres; National AI Centre retains the guidance function.
- What kind of body
- national government department + AI safety institute issuing voluntary guidance
- How it regulates AI
- Voluntary guidance
- How it treats risk
- voluntary, risk-based guidance; reliance on existing technology-neutral laws; proposed mandatory guardrails for high-risk AI NOT proceeding
- What it covers
- —
- Adopted
- 2025-10-21
- In force from
- —
- What happens next
- 2026-07-15 — Office of AI established in the Department of the Prime Minister and Cabinet. An Australian AI standard is in design, to include mandatory requirements for large AI data centres.
- What the regulator can do
- none (voluntary); no AI-specific regulator or penalties — existing privacy, consumer, online-safety and anti-discrimination laws apply
- Maximum penalty
- none (voluntary)Per Voluntary AI Safety Standard (DISR, published 5 Sep 2024, 10 guardrails); Guidance for AI Adoption (National AI Centre, 21 Oct 2025, six essential practices — consolidates/supersedes the 10 guardrails); National AI Plan (Australian Government, 2 Dec 2025) — Austrade (Australian Trade and Investment Commission), Australian Government — reporting the DISR National AI Plan ↗
- Signed the Council of Europe AI treaty?
- no
- Signed up to the OECD AI Principles?
- yes
- Last verified
- 2026-07-26
Brazil (BR)
South America
- The law or rule
- Projeto de Lei nº 2338/2023 — Marco Legal da Inteligência Artificial
- What kind of rule
- bill / draft law
- Official reference
- PL 2338/2023, aprovado pelo Plenário do Senado Federal em 10 Dez 2024; remetido à Câmara dos Deputados em 17 Mar 2025 (situação: tramitação no Senado encerrada; aguardando deliberação na Câmara)
- Does it bind yet?
- Not yet — still a draftStated in PL 2338/2023, aprovado pelo Plenário do Senado Federal em 10 Dez 2024; remetido à Câmara dos Deputados em 17 Mar 2025 (situação: tramitação no Senado encerrada; aguardando deliberação na Câmara) — Senado Federal (senado.leg.br) ↗
- Who enforces it
- Autoridade Nacional de Proteção de Dados (ANPD), proposed to coordinate the Sistema Nacional de Regulação e Governança de IA (SIA); bill under review in the Câmara dos Deputados
- What kind of body
- data-protection authority (proposed SIA coordinator)
- How it regulates AI
- One broad AI law
- How it treats risk
- risk-tiered on the EU model (excessive/prohibited risk and high risk categories), rights of affected persons
- What it covers
- prohibited-practices, high-risk, biometric, automated-decisions
- Adopted
- —
- In force from
- —
- What happens next
- Awaiting vote in the Câmara dos Deputados; if amended, returns to the Senado before presidential sanction
- What the regulator can do
- none yet (draft); as approved by the Senado the bill proposes administrative sanctions coordinated by the SIA/ANPD
- Maximum penalty
- Label explicitly as a PROPOSED penalty in PL 2338/2023 as approved by the Senate on 2024-12-10 — not in force. The bill remains 'Aguardando Parecer' in the Chamber of Deputies, last tramitação 2026-06-17, with no plenary vote (disregard secondary reports of a 27 May 2026 vote).Per PL 2338/2023, aprovado pelo Plenário do Senado Federal em 10 Dez 2024; remetido à Câmara dos Deputados em 17 Mar 2025 (situação: tramitação no Senado encerrada; aguardando deliberação na Câmara) — Senado Federal (senado.leg.br) ↗
- Signed the Council of Europe AI treaty?
- no
- Signed up to the OECD AI Principles?
- yes
- Last verified
- 2026-07-26
Canada (CA)
Northern America
- The law or rule
- Voluntary Code of Conduct on the Responsible Development and Management of Advanced Generative AI Systems
- What kind of rule
- voluntary code of conduct + national strategy
- Official reference
- Voluntary Code of Conduct on the Responsible Development and Management of Advanced Generative AI Systems (Sept 2023); Artificial Intelligence and Data Act (AIDA), part of Bill C-27, died on the prorogation of Parliament 6 Jan 2025 (Parliament of Canada, LEGISinfo, Bill C-27, 44th Parl., 1st Sess.); National AI Strategy 'AI for All' launched 4 Jun 2026
- Does it bind yet?
- No — the bill lapsedStated in Voluntary Code of Conduct on the Responsible Development and Management of Advanced Generative AI Systems (Sept 2023); Artificial Intelligence and Data Act (AIDA), part of Bill C-27, died on the prorogation of Parliament 6 Jan 2025 (Parliament of Canada, LEGISinfo, Bill C-27, 44th Parl., 1st Sess.); National AI Strategy 'AI for All' launched 4 Jun 2026 — Innovation, Science and Economic Development Canada (ISED) ↗
- Who enforces it
- Innovation, Science and Economic Development Canada (ISED); Minister of Artificial Intelligence and Digital Innovation
- What kind of body
- national ministry (no dedicated statutory AI regulator; binding bill lapsed)
- How it regulates AI
- Voluntary guidance
- How it treats risk
- principles-based / voluntary (safety, fairness, transparency, human oversight); strategy is promotion-first
- What it covers
- GPAI, synthetic-media-labeling
- Adopted
- 2023-09-27
- In force from
- —
- What happens next
- No AI-specific statute; AIDA lapsed with Bill C-27 (44th Parliament, 1st Session, never received Royal Assent, last event 2023-04-24). In the 45th Parliament, 1st Session, AI transparency provisions for automated decision tools are before Parliament within privacy Bill C-36 (at second reading, latest event 2026-06-15); Bill C-34 Safe Social Media Act is at second reading (latest event 2026-06-10). Neither is an AI act and neither has passed.
- What the regulator can do
- none (voluntary code; no binding enforcement authority after AIDA lapsed)
- Maximum penalty
- none (voluntary)Per Voluntary Code of Conduct on the Responsible Development and Management of Advanced Generative AI Systems (Sept 2023); Artificial Intelligence and Data Act (AIDA), part of Bill C-27, died on the prorogation of Parliament 6 Jan 2025 (Parliament of Canada, LEGISinfo, Bill C-27, 44th Parl., 1st Sess.); National AI Strategy 'AI for All' launched 4 Jun 2026 — Innovation, Science and Economic Development Canada (ISED) ↗
- Signed the Council of Europe AI treaty?
- signed 11 Feb 2025
- Signed up to the OECD AI Principles?
- yes
- Last verified
- 2026-07-26
Chile (CL)
South America
- The law or rule
- Proyecto de ley que regula los sistemas de inteligencia artificial (Boletín 16821-19)
- What kind of rule
- bill / draft law
- Official reference
- Boletines refundidos 15869-19 (matriz) y 16821-19, 'Regula los sistemas de inteligencia artificial'; iniciativa Mensaje; ingreso 2024-05-07; cámara de origen Cámara de Diputados.
- Does it bind yet?
- Not yet — still a draftStated in Boletines refundidos 15869-19 (matriz) y 16821-19, 'Regula los sistemas de inteligencia artificial'; iniciativa Mensaje; ingreso 2024-05-07; cámara de origen Cámara de Diputados. — Senado de la República de Chile (senado.cl) ↗
- Who enforces it
- Ministerio de Ciencia, Tecnología, Conocimiento e Innovación (MinCiencia); bill under review in the Senado (Comisión Desafíos del Futuro, Ciencia, Tecnología e Innovación); proposed oversight by the Agencia de Protección de Datos Personales, advised by a Consejo Asesor Técnico de Inteligencia Artificial
- What kind of body
- national ministry (sectoral lead); proposed oversight/enforcement body (data-protection agency) under the bill
- How it regulates AI
- One broad AI law
- How it treats risk
- risk-tiered on the EU model — unacceptable (prohibited), high, limited, and no-evident-risk categories
- What it covers
- prohibited-practices, high-risk, automated-decisions
- Adopted
- —
- In force from
- —
- What happens next
- Approved in first constitutional reading by the Cámara de Diputados on 2025-10-13; now in Segundo trámite constitucional (Senado), at 'Primer informe de comisión de Desafíos del Futuro, Ciencia, Tecnología e Innovación'; ESTADO 'En tramitación'; executive urgency 'Simple', last renewed 2026-07-07.
- What the regulator can do
- none yet (draft); the bill proposes administrative sanctions enforced by the Agencia de Protección de Datos Personales, advised by a Consejo Asesor Técnico de Inteligencia Artificial, for prohibited/high-risk violations
- Maximum penalty
- none in force (draft); the pending text (as approved by the Cámara de Diputados) proposes graduated administrative fines, reported by the Senate source at 5,000–20,000 UTM (Unidades Tributarias Mensuales) depending on severity, for unacceptable-risk (prohibited) and high-risk violationsPer Boletines refundidos 15869-19 (matriz) y 16821-19, 'Regula los sistemas de inteligencia artificial'; iniciativa Mensaje; ingreso 2024-05-07; cámara de origen Cámara de Diputados. — Senado de la República de Chile (senado.cl) ↗
- Signed the Council of Europe AI treaty?
- no
- Signed up to the OECD AI Principles?
- yes
- Last verified
- 2026-07-26
China (CN)
Eastern Asia
- The law or rule
- Interim Measures for the Management of Generative AI Services (生成式人工智能服务管理暂行办法), in force 15 August 2023; and Interim Measures for the Administration of AI Anthropomorphic Interactive Services (人工智能拟人化互动服务管理暂行办法), announced 10 April 2026 and in force 15 July 2026.
- What kind of rule
- administrative measures (departmental regulations)
- Official reference
- Interim Measures for the Management of Generative AI Services (生成式人工智能服务管理暂行办法), CAC Order No. 15 (国家互联网信息办公室令第15号), promulgated 10 Jul 2023, effective 15 Aug 2023, Art. 21; Measures for Labeling of AI-Generated Synthetic Content (人工智能生成合成内容标识办法), issued 14 Mar 2025, effective 1 Sep 2025
- Does it bind yet?
- Yes — in forceStated in Interim Measures for the Management of Generative AI Services (生成式人工智能服务管理暂行办法), CAC Order No. 15 (国家互联网信息办公室令第15号), promulgated 10 Jul 2023, effective 15 Aug 2023, Art. 21; Measures for Labeling of AI-Generated Synthetic Content (人工智能生成合成内容标识办法), issued 14 Mar 2025, effective 1 Sep 2025 — Cyberspace Administration of China (CAC) ↗
- Who enforces it
- For the 2026 anthropomorphic-interaction measures: CAC + National Development and Reform Commission + Ministry of Industry and Information Technology + Ministry of Public Security + State Administration for Market Regulation (SAMR replaces the education/science/broadcast ministries that co-issued the 2023 measures).
- What kind of body
- sectoral regulators (national cyberspace/internet regulator + sectoral ministries)
- How it regulates AI
- Through existing sector rules
- How it treats risk
- sectoral; 'inclusive and prudent' classified/tiered supervision (包容审慎、分类分级监管) with algorithm filing and security assessment for public-facing generative services
- What it covers
- GPAI, synthetic-media-labeling
- Adopted
- 2023-07-10
- In force from
- 2023-08-15
- What happens next
- 2026-07-15 — AI Anthropomorphic Interactive Services Measures entered into force. A comprehensive national AI Law is listed in the State Council 2026 Legislative Work Plan as under drafting
- What the regulator can do
- administrative measures (warning, public criticism, order to correct, suspension of service) under Art. 21, plus penalties under referenced statutes (Cybersecurity Law, Data Security Law, PIPL); algorithm filing and security assessment obligations
- Maximum penalty
- Under the 2026 AI Anthropomorphic Interactive Services Measures (Art. 30): warning, public criticism or an order to rectify; RMB 10,000–100,000 for serious non-compliance; and RMB 100,000–200,000, with possible suspension of the service, where public health and safety are endangered. These bands apply to those Measures, not to the 2023 generative-AI measures.Per Interim Measures for the Management of Generative AI Services (生成式人工智能服务管理暂行办法), CAC Order No. 15 (国家互联网信息办公室令第15号), promulgated 10 Jul 2023, effective 15 Aug 2023, Art. 21; Measures for Labeling of AI-Generated Synthetic Content (人工智能生成合成内容标识办法), issued 14 Mar 2025, effective 1 Sep 2025 — Cyberspace Administration of China (CAC) ↗
- Signed the Council of Europe AI treaty?
- no
- Signed up to the OECD AI Principles?
- no
- Last verified
- 2026-07-26
European Union (EU)
Europe
- The law or rule
- Artificial Intelligence Act — Regulation (EU) 2024/1689
- What kind of rule
- regulation
- Official reference
- Regulation (EU) 2024/1689 of 13 June 2024 (Artificial Intelligence Act), OJ L, 2024/1689, 12.7.2024, CELEX 32024R1689, ELI http://data.europa.eu/eli/reg/2024/1689/oj — as amended by Regulation (EU) 2026/1744 of 8 July 2026 (Digital Omnibus on AI), OJ L, 2026/1744, 24.7.2026, CELEX 32026R1744, in force 27 July 2026. Corrigenda 32024R1689R(01)–R(04) are also recorded on EUR-Lex.
- Does it bind yet?
- Partly — phasing inStated in Regulation (EU) 2024/1689 of 13 June 2024 (Artificial Intelligence Act) — EUR-Lex — Publications Office of the European Union (Official Journal text) ↗
- Who enforces it
- National competent authorities designated by each Member State (Art. 70), the European AI Office (Art. 64) and the European AI Board (Art. 65); the Commission has exclusive powers over general-purpose AI models under Chapter V (Art. 88(1)). From 27 July 2026, Arts. 75a–75d inserted by Regulation (EU) 2026/1744 give the AI Office direct supervisory and enforcement powers, including fines applying Art. 99(3)–(7), over certain AI systems within its competence.
- What kind of body
- dedicated AI authority
- How it regulates AI
- One broad AI law
- How it treats risk
- risk-tiered (prohibited / high-risk / limited / minimal) plus a distinct general-purpose AI (GPAI) regime
- What it covers
- prohibited-practices, high-risk, GPAI, biometric, synthetic-media-labeling
- Adopted
- 2024-06-13
- In force from
- 2024-08-01
- What happens next
- General application date 2 August 2026 (Art. 113, second paragraph), subject to the carve-outs in Art. 113, third paragraph. Chapters I and II have applied since 2 February 2025; Chapter III Section 4, Chapters V, VII and XII and Art. 78 since 2 August 2025 (except Art. 101). Regulation (EU) 2026/1744 (Digital Omnibus on AI), OJ L, 2026/1744, 24.7.2026, in force from 27 July 2026, replaces Art. 113, third paragraph, point (c): Chapter III, Sections 1, 2 and 3 (except Art. 6(5)) apply from 2 December 2027 for Art. 6(2)/Annex III high-risk systems and from 2 August 2028 for Art. 6(1)/Annex I high-risk systems. New point (d): Articles 102 to 110 apply from 27 July 2026. Amended point (a): new Art. 5(1) points (ba) and (bb) and Art. 5(1a)/(1b) apply from 2 December 2026. New Art. 111(4): providers of AI systems generating synthetic audio, image, video or text placed on the market before 2 August 2026 must comply with Art. 50(2) by 2 December 2026. Art. 111(2) as replaced retains 2 August 2030 for high-risk systems intended for use by public authorities.
- What the regulator can do
- administrative fines + market-surveillance
- Maximum penalty
- Art. 99(3): up to EUR 35 000 000 or, for an undertaking, up to 7 % of total worldwide annual turnover for the preceding financial year, whichever is higher (Art. 5 prohibitions). Art. 99(4): up to EUR 15 000 000 or 3 %, whichever is higher, for a closed list of provisions (Arts. 16, 22, 23, 24, 26, 31, 33(1),(3),(4), 34, 50 — plus new point (da), Art. 25(2) and (4), from 27 July 2026). Art. 99(5): up to EUR 7 500 000 or 1 %, whichever is higher, for incorrect/incomplete/misleading information. Art. 99(6): SMEs including start-ups pay the lower of the percentage or the amount; new Art. 99(6a) extends the lower-of rule to small mid-cap enterprises from 27 July 2026. Separately Art. 101: Commission fines for GPAI model providers up to 3 % of worldwide turnover or EUR 15 000 000, whichever is higher.Per Regulation (EU) 2024/1689 of 13 June 2024 (Artificial Intelligence Act), OJ L, 2024/1689, 12.7.2024, CELEX 32024R1689, ELI http://data.europa.eu/eli/reg/2024/1689/oj — as amended by Regulation (EU) 2026/1744 of 8 July 2026 (Digital Omnibus on AI), OJ L, 2026/1744, 24.7.2026, CELEX 32026R1744, in force 27 July 2026. Corrigenda 32024R1689R(01)–R(04) are also recorded on EUR-Lex. — EUR-Lex — Publications Office of the European Union (Official Journal text) ↗
- Signed the Council of Europe AI treaty?
- ratified 15 May 2026
- Signed up to the OECD AI Principles?
- yes
- Last verified
- 2026-07-26
Israel (IL)
Western Asia
- The law or rule
- Responsible Innovation: Israel's Policy on Artificial Intelligence Regulation and Ethics
- What kind of rule
- national policy / non-binding guidance
- Official reference
- Responsible Innovation: Israel's Policy on Artificial Intelligence Regulation and Ethics (Ministry of Innovation, Science and Technology & Ministry of Justice, December 2023)
- Does it bind yet?
- No — voluntaryStated in Responsible Innovation: Israel's Policy on Artificial Intelligence Regulation and Ethics (Ministry of Innovation, Science and Technology & Ministry of Justice, December 2023) — State of Israel — Ministry of Innovation, Science and Technology & Ministry of Justice (gov.il) ↗
- Who enforces it
- Ministry of Innovation, Science and Technology with the Ministry of Justice (Office of Legal Counsel and Legislative Affairs, Economic Law Department); implementation via sectoral regulators
- What kind of body
- national ministries + sectoral regulators
- How it regulates AI
- Voluntary guidance
- How it treats risk
- sectoral, risk-based, principles-based (soft regulation; sector-specific regulators over horizontal legislation)
- What it covers
- automated-decisions
- Adopted
- 2023-12-17 — the gov.il record displays 'Publish Date: 17.12.2023' and the body reads 'Jerusalem 17th December 2023'. Secondary trackers give 13 December 2023 (ministerial endorsement); match the authority.
- In force from
- —
- What happens next
- Sectoral regulators to develop tailored, risk-based rules; AI policy knowledge/coordination centre; periodic reassessment of the need for horizontal legislation
- What the regulator can do
- none (voluntary)
- Maximum penalty
- none (voluntary)Per Responsible Innovation: Israel's Policy on Artificial Intelligence Regulation and Ethics (Ministry of Innovation, Science and Technology & Ministry of Justice, December 2023) — State of Israel — Ministry of Innovation, Science and Technology & Ministry of Justice (gov.il) ↗
- Signed the Council of Europe AI treaty?
- signed 5 Sep 2024
- Signed up to the OECD AI Principles?
- yes
- Last verified
- 2026-07-26
Japan (JP)
Eastern Asia
- The law or rule
- Act on the Promotion of Research and Development and Utilization of Artificial Intelligence-related Technology (AI Promotion Act)
- What kind of rule
- act/statute (basic/promotion act)
- Official reference
- Act No. 53 of 2025 (令和7年法律第53号) — Act on Promotion of Research and Development and Utilization of AI-Related Technologies (人工知能関連技術の研究開発及び活用の推進に関する法律); e-Gov law ID 507AC0000000053.
- Does it bind yet?
- Yes — in forceStated in Act No. 53 of 2025 (令和7年法律第53号) — Government of Japan — Japanese Law Translation (Ministry of Justice) official translation ↗
- Who enforces it
- Cabinet — Artificial Intelligence Strategy Headquarters (chaired by the Prime Minister); Minister for AI Strategy
- What kind of body
- dedicated AI authority (cross-government strategic headquarters, promotion-oriented)
- How it regulates AI
- One broad AI law
- How it treats risk
- promotion-first / principles-based — sets basic policy, an AI Basic Plan and ministry guidance; no prohibitions or penalties; investigation and cooperation-request mechanisms only
- What it covers
- —
- Adopted
- 2025-06-04
- In force from
- 2025-06-04 (promulgated; partial entry into force) / 2025-09-01 (full entry into force, together with the provisions establishing the AI Strategy Headquarters)
- What happens next
- 2026-07-14 — Phase II AI Basic Plan (人工知能基本計画（第Ⅱ期）) adopted by Cabinet decision; 2026-07-10 — 5th meeting of the AI Strategy Headquarters. First AI Basic Plan Cabinet-decided 2025-12-23 (Reiwa 7 = 2025).
- What the regulator can do
- none (no penalties); government may conduct investigations/information-gathering and request cooperation from operators; guidance-based
- Maximum penalty
- None — the Act is a promotion statute and contains no penalty provisions. State this explicitly rather than leaving the cell blank, which reads as unresearched.Per Act No. 53 of 2025 (令和7年法律第53号) — Act on Promotion of Research and Development and Utilization of AI-Related Technologies (人工知能関連技術の研究開発及び活用の推進に関する法律); e-Gov law ID 507AC0000000053. — Government of Japan — Japanese Law Translation (Ministry of Justice) official translation ↗
- Signed the Council of Europe AI treaty?
- signed 11 Feb 2025
- Signed up to the OECD AI Principles?
- yes
- Last verified
- 2026-07-26
Nigeria (NG)
Sub-Saharan Africa
- The law or rule
- Draft National Artificial Intelligence Strategy, August 2024 The document hosted on the official NCAIR/NITDA domain is marked 'DRAFT AUGUST, 2024' and every page header reads 'DRAFT NATIONAL ARTIFICIAL INTELLIGENCE STRATEGY 2024'. Owner: Federal Ministry of Communication, Innovation and Digital Economy, with NITDA and others.
- What kind of rule
- national strategy (draft)
- Official reference
- National Artificial Intelligence Strategy (NAIS), Draft, August 2024 (Federal Ministry of Communications, Innovation & Digital Economy with NITDA/NCAIR)
- Does it bind yet?
- No — voluntaryStated in National Artificial Intelligence Strategy (NAIS), Draft, August 2024 (Federal Ministry of Communications, Innovation & Digital Economy with NITDA/NCAIR) — National Centre for Artificial Intelligence and Robotics (NCAIR), NITDA (ncair.nitda.gov.ng) ↗
- Who enforces it
- National Information Technology Development Agency (NITDA) / National Centre for Artificial Intelligence and Robotics (NCAIR), under the Federal Ministry of Communications, Innovation and Digital Economy (FMCIDE)
- What kind of body
- national ICT/AI agency (strategy body)
- How it regulates AI
- National strategy only
- How it treats risk
- promotion-first national strategy; proposes (under the governance pillar) National AI Principles, an AI Governance Regulatory Body, a comprehensive risk-management framework and a High-Level AI Ethics Expert Group (AIEEG)/National AI Ethics Commission — proposals, not enacted
- What it covers
- —
- Adopted
- 2024-08
- In force from
- —
- What happens next
- National Digital Economy and E-Governance Bill — reported passed by the National Assembly and awaiting presidential assent as at 26 July 2026; it would give NITDA statutory authority over algorithms and digital platforms. It is not yet law, and its stage could not be confirmed against a primary source. This is a different instrument from the NIMC Act 2026, signed in June 2026.
- What the regulator can do
- none (voluntary)
- Maximum penalty
- none (voluntary)Per National Artificial Intelligence Strategy (NAIS), Draft, August 2024 (Federal Ministry of Communications, Innovation & Digital Economy with NITDA/NCAIR) — National Centre for Artificial Intelligence and Robotics (NCAIR), NITDA (ncair.nitda.gov.ng) ↗
- Signed the Council of Europe AI treaty?
- no
- Signed up to the OECD AI Principles?
- no
- Last verified
- 2026-07-26
Saudi Arabia (SA)
Western Asia
- The law or rule
- AI Ethics Principles, listed on SDAIA's Laws and Regulations index as a draft. SDAIA's other listed AI instruments are guidance — the Generative AI Guideline, Deepfakes Guidelines, AI Adoption Framework, and occupational and academic frameworks. The only binding instrument in that index is the Personal Data Protection Law and its Implementing Regulation, which govern personal data rather than AI specifically.
- What kind of rule
- voluntary ethics framework
- Official reference
- SDAIA AI Ethics Principles, September 2023, Version 1.0, issued under Council of Ministers' Resolution No. 292 dated 27/04/1441 AH (Art. 10(1)); seven principles
- Does it bind yet?
- No — voluntaryStated in SDAIA AI Ethics Principles, September 2023, Version 1.0, issued under Council of Ministers' Resolution No. 292 dated 27/04/1441 AH (Art. 10(1)); seven principles — Saudi Data and Artificial Intelligence Authority (SDAIA) ↗
- Who enforces it
- Saudi Data and Artificial Intelligence Authority (SDAIA)
- What kind of body
- dedicated data & AI authority
- How it regulates AI
- Voluntary guidance
- How it treats risk
- risk-tiered (little/no, limited, high, unacceptable) + seven ethics principles
- What it covers
- high-risk, automated-decisions
- Adopted
- 2023-09
- In force from
- —
- What happens next
- None announced by SDAIA.
- What the regulator can do
- none (voluntary)
- Maximum penalty
- none (voluntary)Per SDAIA AI Ethics Principles, September 2023, Version 1.0, issued under Council of Ministers' Resolution No. 292 dated 27/04/1441 AH (Art. 10(1)); seven principles — Saudi Data and Artificial Intelligence Authority (SDAIA) ↗
- Signed the Council of Europe AI treaty?
- no
- Signed up to the OECD AI Principles?
- no
- Last verified
- 2026-07-26
Singapore (SG)
South-Eastern Asia
- The law or rule
- Model AI Governance Framework for Agentic AI (IMDA), launched 2026-01-22, updated May 2026. List as an addition, not a replacement — IMDA does not state a supersession relationship. Retain the 2024 Generative AI framework and AI Verify.
- What kind of rule
- voluntary framework / guidance
- Official reference
- Model AI Governance Framework for Generative AI (IMDA & AI Verify Foundation, 30 May 2024); AI Verify (2023); Model AI Governance Framework 1st ed. 2019, 2nd ed. Jan 2020 (PDPC)
- Does it bind yet?
- non-binding at national level; sector regulators impose binding obligations that reach AI, notably Monetary Authority of Singapore requirements for financial institutionsStated in Model AI Governance Framework for Generative AI (IMDA & AI Verify Foundation, 30 May 2024); AI Verify (2023); Model AI Governance Framework 1st ed. 2019, 2nd ed. Jan 2020 (PDPC) — Infocomm Media Development Authority (IMDA) & AI Verify Foundation ↗
- Who enforces it
- Infocomm Media Development Authority (IMDA) and Personal Data Protection Commission (PDPC); AI Verify Foundation
- What kind of body
- sectoral regulator (infocomm-media / data-protection) issuing voluntary frameworks
- How it regulates AI
- Voluntary guidance
- How it treats risk
- principles-based and voluntary; nine dimensions in the GenAI framework (accountability, data, trusted development/deployment, incident reporting, testing & assurance, security, content provenance, safety & alignment R&D, AI for public good)
- What it covers
- GPAI, synthetic-media-labeling
- Adopted
- 2024-05-30
- In force from
- —
- What happens next
- Ongoing extensions to the voluntary framework (incl. work on generative/agentic AI); no move to binding AI-specific legislation announced (date —)
- What the regulator can do
- none (voluntary); assurance via AI Verify testing and self-assessment
- Maximum penalty
- none (voluntary)Per Model AI Governance Framework for Generative AI (IMDA & AI Verify Foundation, 30 May 2024); AI Verify (2023); Model AI Governance Framework 1st ed. 2019, 2nd ed. Jan 2020 (PDPC) — Infocomm Media Development Authority (IMDA) & AI Verify Foundation ↗
- Signed the Council of Europe AI treaty?
- no
- Signed up to the OECD AI Principles?
- yes
- Last verified
- 2026-07-26
South Korea (Republic of Korea) (KR)
Eastern Asia
- The law or rule
- Basic Act on the Development of Artificial Intelligence and the Establishment of a Foundation for Trust (AI Basic Act / AI Framework Act)
- What kind of rule
- act/statute (framework act)
- Official reference
- Act No. 20676 (법률 제20676호), promulgated 21 January 2025, in force 22 January 2026; Enforcement Decree, Presidential Decree No. 36053 (대통령령 제36053호), 21 January 2026, in force 22 January 2026.
- Does it bind yet?
- Yes — in forceStated in Act No. 20676 (법률 제20676호), promulgated 21 January 2025, in force 22 January 2026; Enforcement Decree, Presidential Decree No. 36053 (대통령령 제36053호), 21 January 2026, in force 22 January 2026. — Ministry of Science and ICT (MSIT) ↗
- Who enforces it
- Ministry of Science and ICT (MSIT)
- What kind of body
- national ministry (science/ICT) designated as AI competent authority
- How it regulates AI
- One broad AI law
- How it treats risk
- risk-based — obligations focused on 'high-impact AI' plus generative-AI transparency; principles-based development and trust duties
- What it covers
- GPAI, high-risk, synthetic-media-labeling
- Adopted
- 2025-01-21
- In force from
- 2026-01-22
- What happens next
- 2026-07-21 — deferred Framework Act provisions (public-sector AI adoption, AI research institute, AI-vulnerable-group access and cost support, public data for training data) and the amended Enforcement Decree entered into force (Decree cleared the State Council 2026-07-14 after legislative notice 2026-05-21). The 21 July 2026 commencement date is as stated by the Ministry of Science and ICT (MSIT); it has not been read off the statute record on law.go.kr, so it is reported here on MSIT's authority rather than on the primary legislative record.
- What the regulator can do
- administrative fines + cease/corrective orders and fact-finding investigations (MSIT); criminal penalty for unlawful disclosure of confidential information
- Maximum penalty
- Administrative fine up to KRW 30,000,000 (Art. 43); criminal — up to 3 years' imprisonment or a fine up to KRW 30,000,000 for unlawful disclosure/misuse of confidential information (Art. 42); ~1-year grace on administrative finesPer Act No. 20676 (법률 제20676호), promulgated 21 January 2025, in force 22 January 2026; Enforcement Decree, Presidential Decree No. 36053 (대통령령 제36053호), 21 January 2026, in force 22 January 2026. — Ministry of Science and ICT (MSIT) ↗
- Signed the Council of Europe AI treaty?
- no
- Signed up to the OECD AI Principles?
- yes
- Last verified
- 2026-07-26
United Kingdom (GB)
Europe
- The law or rule
- A pro-innovation approach to AI regulation (2023 White Paper) — five non-statutory cross-sector principles delivered through existing regulators; Data (Use and Access) Act 2025
- What kind of rule
- voluntary framework (non-statutory principles) + statute (Data (Use and Access) Act 2025)
- Official reference
- A pro-innovation approach to AI regulation — DSIT White Paper (CP 815, 29 Mar 2023) and Government response (CP 1019, 6 Feb 2024); Data (Use and Access) Act 2025 (c. 18), Royal Assent 19 Jun 2025
- Does it bind yet?
- No — voluntaryStated in A pro-innovation approach to AI regulation — GOV.UK — Department for Science, Innovation and Technology (DSIT) ↗
- Who enforces it
- Department for Science, Innovation and Technology (DSIT); existing sector regulators (ICO, Ofcom, FCA, CMA, MHRA, etc.); AI Security Institute (formerly AI Safety Institute, renamed 14 Feb 2025)
- What kind of body
- sectoral regulators
- How it regulates AI
- Through existing sector rules
- How it treats risk
- principles-based, context- and sector-specific (five cross-sector principles, non-statutory)
- What it covers
- —
- Adopted
- 2023-03-29
- In force from
- —
- What happens next
- No AI Bill in the 2026 session — the King's Speech 2026 background briefing notes contain no AI bill or measure. The Regulating for Growth Bill (Department for Business and Trade), announced in the King's Speech of May 2026, is a cross-cutting bill creating statutory sandbox and regulatory-innovation powers that can be applied to AI among other sectors; the Department published its own guidance on those powers, Regulatory Sandboxes: What Businesses Can Expect, on 8 July 2026. It is not an AI-specific instrument and is not the United Kingdom's AI law. Background chain: March 2023 white paper, February 2024 government response, 2025-10-20 DSIT blueprint for AI regulation centred on the AI Growth Lab. Competent authorities remain the existing sectoral regulators (ICO, Ofcom, FCA, CMA).
- What the regulator can do
- none (voluntary); AI risks addressed through existing sector regulators' own statutory powers
- Maximum penalty
- none (voluntary)Per A pro-innovation approach to AI regulation — DSIT White Paper (CP 815, 29 Mar 2023) and Government response (CP 1019, 6 Feb 2024); Data (Use and Access) Act 2025 (c. 18), Royal Assent 19 Jun 2025 — GOV.UK — Department for Science, Innovation and Technology (DSIT) ↗
- Signed the Council of Europe AI treaty?
- signed 5 Sep 2024
- Signed up to the OECD AI Principles?
- yes
- Last verified
- 2026-07-26
United States — Colorado (US-CO)
Northern America
- The law or rule
- SB 26-189, Automated Decision-Making Technology, which repeals and reenacts with amendments part 17 of article 1 of title 6 C.R.S. Status: Became Law. SB 24-205 is REPEALED and never applied to anyone.
- What kind of rule
- act/statute (state law)
- Official reference
- C.R.S. title 6, article 1, part 17 (§§ 6-1-1701 to 6-1-1706), as repealed and reenacted with amendments by SB 26-189 (2026).
- Does it bind yet?
- Adopted, but not applying yetStated in C.R.S. title 6, article 1, part 17 (§§ 6-1-1701 to 6-1-1706), as repealed and reenacted with amendments by SB 26-189 (2026). — Colorado General Assembly (leg.colorado.gov) ↗
- Who enforces it
- Colorado Attorney General (Department of Law)
- What kind of body
- sub-national attorney general (exclusive enforcer)
- How it regulates AI
- One broad AI law
- How it treats risk
- focus on consequential decisions made by automated decision-making technology (education, employment, housing, financial services, insurance, healthcare, government benefits)
- What it covers
- high-risk, automated-decisions
- Adopted
- 2026-05-14 (SB 26-189 signed by the Governor; Senate concurrence 34-1 on 2026-05-12).
- In force from
- 2027-01-01 for the act generally; §§ 6-1-1704(4), 6-1-1705(3), 6-1-1706(6), 10-3-1104.9(3)(e) and sections 4–6 took effect upon passage, 2026-05-14. Use the ENROLLED ACT TEXT (Section 5), not the bill-page summary field, which renders an incorrect 2026-08-12 effective date.
- What happens next
- On or before 2027-01-01: Attorney General must adopt implementing rules. 2027-01-01: the act applies to consequential decisions made on or after that date, and developer technical-documentation duties begin. From January 2028 and annually: AG enforcement reporting at the SMART Act hearing.
- What the regulator can do
- administrative — exclusive enforcement by the Colorado Attorney General as a deceptive trade practice under the Colorado Consumer Protection Act; no private right of action; a 60-day notice-and-cure requirement applies before enforcement until 1 January 2030
- Maximum penalty
- SB 26-189 sets no penalty figure of its own. A violation is a deceptive trade practice enforceable exclusively by the Colorado Attorney General under the Colorado Consumer Protection Act, subject to a 60-day right to cure and with no private right of action; civil penalties are those of the Consumer Protection Act at C.R.S. § 6-1-112. The specific figure is not restated here because it could not be confirmed against the official C.R.S. text.Per C.R.S. title 6, article 1, part 17 (§§ 6-1-1701 to 6-1-1706), as repealed and reenacted with amendments by SB 26-189 (2026). — Colorado General Assembly (leg.colorado.gov) ↗
- Signed the Council of Europe AI treaty?
- n/a
- Signed up to the OECD AI Principles?
- n/a
- Last verified
- 2026-07-26
United States — federal (US)
Northern America
- The law or rule
- Executive Order 14179 (23 January 2025), which remains in force; Executive Order 14365, 'Ensuring a National Policy Framework for Artificial Intelligence' (11 December 2025); and Executive Order 14409, 'Promoting Advanced Artificial Intelligence Innovation and Security' (2 June 2026). No comprehensive federal AI statute has been enacted.
- What kind of rule
- executive policy plan / national strategy (non-binding) + OMB agency guidance
- Official reference
- EO 14365, 90 FR 58499 (16 Dec 2025); EO 14409, 91 FR 34565 (5 Jun 2026); EO 14179 (23 Jan 2025).
- Does it bind yet?
- non-binding on private developers; no federal AI statute has been enacted, and the obligations that exist are directed at federal agencies through executive ordersStated in EO 14365, 90 FR 58499 (16 Dec 2025); EO 14409, 91 FR 34565 (5 Jun 2026); EO 14179 (23 Jan 2025). — The White House / Executive Office of the President ↗
- Who enforces it
- Sectoral federal agencies (FTC, EEOC, CFPB, FDA) applying existing law, coordinated with NIST and the OMB / White House Office of Science and Technology Policy
- What kind of body
- sectoral regulators + standards body (no dedicated horizontal AI authority)
- How it regulates AI
- Through existing sector rules
- How it treats risk
- sectoral / innovation-and-deregulation-first; no cross-cutting statutory risk tiers
- What it covers
- GPAI, public-sector-use, synthetic-media-labeling
- Adopted
- 2025-07-23
- In force from
- —
- What happens next
- Deadlines set by EO 14365: AI Litigation Task Force established by the Attorney General within 30 days; Secretary of Commerce evaluation of State AI laws within 90 days; Commerce/NTIA Policy Notice on State eligibility for discretionary funding within 90 days; FTC Chairman policy statement within 90 days; FCC proceeding on a Federal reporting and disclosure standard within 90 days of the section 4 identification; Sec. 8 legislative recommendation for a uniform preemptive Federal framework. White House National Policy Framework for AI issued 2026-03-20.
- What the regulator can do
- none under the framework itself (non-binding); sectoral agencies enforce existing statutes (FTC Act, ECOA, civil-rights and product-safety law)
- Maximum penalty
- none (framework is non-binding; sectoral penalties apply under existing law)Per EO 14365, 90 FR 58499 (16 Dec 2025); EO 14409, 91 FR 34565 (5 Jun 2026); EO 14179 (23 Jan 2025). — The White House / Executive Office of the President ↗
- Signed the Council of Europe AI treaty?
- signed 5 Sep 2024
- Signed up to the OECD AI Principles?
- yes
- Last verified
- 2026-07-26
International instruments — 2 agreements that countries adopt
These are not jurisdictions and bind nobody by themselves. They are agreements and recommendations that individual countries choose to sign up to or write into their own law.
Council of Europe (INTL-COE)
International
- The law or rule
- Framework Convention on Artificial Intelligence and Human Rights, Democracy and the Rule of Law (CETS No. 225)
- What kind of rule
- treaty
- Official reference
- CETS No. 225, opened for signature at Vilnius on 5 September 2024. The Council of Europe Treaty Office also publishes a certified copy of the Convention in its corrected version, in accordance with the Committee of Ministers' decision at the 1527th meeting of the Ministers' Deputies, 30 April 2025.
- Does it bind yet?
- Adopted, but not applying yetStated in CETS No. 225, opened for signature at Vilnius on 5 September 2024. The Council of Europe Treaty Office also publishes a certified copy of the Convention in its corrected version, in accordance with the Committee of Ministers' decision at the 1527th meeting of the Ministers' Deputies, 30 April 2025. — Council of Europe — Treaty Office (chart of signatures and ratifications, CETS 225) ↗
- Who enforces it
- Council of Europe (Secretary General as depositary; Treaty Office); interim custodian Committee CDNET (from 1 Jan 2026); Conference of the Parties upon entry into force
- What kind of body
- intergovernmental organisation
- How it regulates AI
- Binding treaty
- How it treats risk
- principles-based (human rights, democracy and rule of law across the AI lifecycle; graduated risk- and impact-management obligations)
- What it covers
- public-sector-use
- Adopted
- —
- In force from
- —
- What happens next
- Entry into force requires five ratifications including at least three Council of Europe member States; the Treaty Office publishes a per-State chart that stamps its own status date.
- What the regulator can do
- none (no fines; Parties implement through domestic measures; follow-up via the Conference of the Parties)
- Maximum penalty
- none (treaty sets no penalties; remedies and sanctions are for Parties' domestic law)Per CETS No. 225, opened for signature at Vilnius on 5 September 2024. The Council of Europe Treaty Office also publishes a certified copy of the Convention in its corrected version, in accordance with the Committee of Ministers' decision at the 1527th meeting of the Ministers' Deputies, 30 April 2025. — Council of Europe — Treaty Office (chart of signatures and ratifications, CETS 225) ↗
- Signed the Council of Europe AI treaty?
- n/a
- Signed up to the OECD AI Principles?
- n/a
- Last verified
- 2026-07-26
OECD (INTL-OECD)
International
- The law or rule
- Recommendation of the Council on Artificial Intelligence (OECD AI Principles)
- What kind of rule
- recommendation (non-binding OECD Council act)
- Official reference
- Recommendation of the Council on Artificial Intelligence, OECD/LEGAL/0449 (adopted 22 May 2019; amended 3 May 2024)
- Does it bind yet?
- in force as a Recommendation, and not legally binding on Adherents — the status the OECD itself records for the instrumentStated in Recommendation of the Council on Artificial Intelligence, OECD/LEGAL/0449 (adopted 22 May 2019; amended 3 May 2024) — OECD — OECD.AI Policy Observatory ↗
- Who enforces it
- Organisation for Economic Co-operation and Development (OECD); OECD.AI Policy Observatory; Committee on Digital Economy Policy
- What kind of body
- intergovernmental organisation
- How it regulates AI
- Voluntary guidance
- How it treats risk
- principles-based (values-based principles for trustworthy AI plus policy recommendations to governments)
- What it covers
- —
- Adopted
- Adopted 22 May 2019; amended 3 May 2024. Instrument reference OECD/LEGAL/0449.
- In force from
- —
- What happens next
- —
- What the regulator can do
- none (voluntary); adherence monitored, no sanctions
- Maximum penalty
- none (voluntary)Per Recommendation of the Council on Artificial Intelligence, OECD/LEGAL/0449 (adopted 22 May 2019; amended 3 May 2024) — OECD — OECD.AI Policy Observatory ↗
- Signed the Council of Europe AI treaty?
- n/a
- Signed up to the OECD AI Principles?
- n/a (is the OECD)
- Last verified
- 2026-07-26
How this register is maintained
- Every row names the law or rule, who enforces it, the official reference, and a live link to the source it comes from.
- Whether a rule binds you yet, and the maximum penalty, are stated exactly as the official text provides them — with the reference and link shown in the same row. We show what the source says; we do not assert it ourselves.
- Every row was re-checked against its official source on 26 July 2026. Where a detail could not be confirmed against a primary source, it was removed rather than published — and where a whole country's position could not be confirmed, that country was held back from this release rather than shown unverified.
- Each row carries its own last-checked date. A change — a law coming into force, a bill lapsing — is recorded row by row, not assumed.
- Rows are listed alphabetically within each group. The register does not rank countries or advise which rules to follow.
- Informational only — not legal, regulatory or compliance advice. Check the official text before relying on anything here.
Is the EU Artificial Intelligence Act in force?
The EU AI Act (Regulation (EU) 2024/1689) entered into force on 1 August 2024 and applies in phases: the prohibited-practices rules from 2 February 2025, general-purpose AI model obligations from 2 August 2025, and the general body of obligations from 2 August 2026 — except most of the high-risk rules, which Regulation (EU) 2026/1744 sets to apply from 2 December 2027 for Annex III systems and from 2 August 2028 for Annex I product-embedded systems. Its status in this register is therefore "partially-in-force/phased".
Article 113 of Regulation (EU) 2024/1689, as amended by Regulation (EU) 2026/1744, sets the phased application dates. The amending act, the Digital Omnibus on AI, is Regulation (EU) 2026/1744, published in OJ L, 2026/1744 of 24 July 2026 and in force from 27 July 2026; it amends Regulation (EU) 2024/1689, including Article 113. The second paragraph of Article 113 — general application from 2 August 2026 — is unamended, and the dates the amendment sets are recorded article by article in the EU row. Maximum administrative fines under Article 99(3) reach EUR 35,000,000 or 7% of total worldwide annual turnover, whichever is higher.
Is the Council of Europe's AI Convention in force?
No. The 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 but has not entered into force. Entry into force requires five ratifications including at least three Council of Europe member states; as of 13 July 2026 the European Union is the sole ratifying party (15 May 2026). This register therefore records it as adopted, but not applying yet.
The Convention's entry-into-force condition and the current signature and ratification status are recorded on the Council of Europe Treaty Office chart for CETS No. 225, which states the status as of the date it is viewed. The treaty sets out obligations for parties; it provides no penalties of its own.
Source: Council of Europe Treaty Office — Chart of signatures and ratifications, CETS No. 225 ↗
How current is the information in this register?
Each row is verified against its official source and carries its own last-verified date. Because AI laws and their status change frequently and differ by jurisdiction, currency is recorded per row rather than for the register as a whole.
A change of legal status is recorded only when the official instrument confirms it. For example, the EU Digital Omnibus on AI (procedure 2025/0359(COD)) was approved by the European Parliament on 16 June 2026, adopted by the Council on 29 June 2026, signed on 8 July 2026 and published as Regulation (EU) 2026/1744 in OJ L, 2026/1744 of 24 July 2026, in force from 27 July 2026; the register updated the affected European Union cells on publication, and each date the Regulation sets is recorded against the article that carries it.
Source: European Parliament Legislative Observatory — procedure file 2025/0359(COD) ↗
Cite this dataset
1BusinessWorld AI Center, "Global AI Governance Register." https://1businessworld.com/ai-center/global-ai-governance-register/ Version as of July 28, 2026.
The AI Center is informational only. It is provided by 1BusinessWorld strictly for general informational and educational purposes. Nothing in the AI Center constitutes, or should be construed as, legal, regulatory, compliance, technical, engineering, security, investment, financial, or other professional advice, or a recommendation, endorsement, solicitation, or offer regarding any technology, product, model, provider, framework, or course of action. 1BusinessWorld is not a law firm, regulatory authority, standards body, conformity-assessment or certification body, or investment adviser, and nothing in the AI Center creates any advisory, fiduciary, attorney-client, or other professional relationship with 1BusinessWorld. Although the AI Center references official materials published by legislatures, regulators, standards bodies, research organizations, and other named authorities, 1BusinessWorld makes no representation or warranty, express or implied, as to the accuracy, completeness, timeliness, or fitness for any purpose of any content, and, to the fullest extent permitted by law, disclaims all liability for any loss or damage of any kind arising directly or indirectly from the use of, or reliance on, any information presented. Laws, regulations, standards, technical practices, and AI capabilities change frequently and differ by jurisdiction; readers must verify all information against the current official text or source and consult qualified legal, compliance, technical, and other professional advisors before acting. Any decision relating to the development, deployment, procurement, or governance of AI systems is made solely at the reader's own risk. Last reviewed: July 28, 2026.