Which countries regulate frontier AI and catastrophic risk?
Where the law puts duties, limits or bans on the most capable AI models because of severe or catastrophic risk: safety frameworks, testing, incident reporting, shutdown capability, or a ban on building superintelligence.
3 of 108 reviewed jurisdictions address this in at least one tracked instrument. Updated 2026-11-16.
Each answer cites the instrument it rests on; open it for the provision and the full analysis. Whether a rule is binding law, a bill or guidance comes from the instrument's type and status, not from wording.
“Not found” means none of the instruments we track for that jurisdiction addresses the topic. It is not a finding that no such rule exists.
Yes — binding law in force (1)
South Korea
Yes — binding law in forceAI operators above a prescribed cumulative training-compute threshold must implement AI safety measures.
- South Korea AI Trust Framework Act, 제32조제1항 — Act, In Force (Amended)
Status verified 2026-10-20
0 of 1 regions address this
- SeoulNot found in the instruments we track
Partly — regional, framework or general (non-AI) law (1)
United States
Partly — regional, framework or general (non-AI) lawCovered by state law. At federal level: proposed only.
Covered providers of extremely compute-intensive AI must maintain shutdown capabilities and report severe loss-of-control incidents.
- US AI Kill Switch Bill for Rogue Systems, Sec. 2220F(b)(1)(A) — Bill, Proposed
- US Comprehensive AI Framework Bill, Sec. 111(g)(1) — Bill, Proposed
- US Artificial Superintelligence Ban Bill, Sections 8-10 — Bill, Proposed
- and 5 more
A possible change to a cited instrument is not yet verified.
Status verified 2026-09-11
5 of 50 states address this
- AlabamaNot found in the instruments we track
- AlaskaNot found in the instruments we track
- ArizonaNot found in the instruments we track
- ArkansasNot found in the instruments we track
- CaliforniaPartly — regional, framework or general (non-AI) law
It imposes safety-framework, assessment, transparency, and incident-reporting duties for compute-threshold frontier models and catastrophic risks.
- California Frontier AI Transparency Act, Business and Professions Code § 22757.11(i)(1) — Act, In Force
Status verified 2026-09-10
- ColoradoNot found in the instruments we track
- ConnecticutPartly — regional, framework or general (non-AI) law
Large frontier developers must maintain anonymous employee reporting for activities posing substantial public safety danger from catastrophic risk.
- Connecticut Online Safety and AI Law, Sec. 2(c)(1) — Act, In Force
Status verified 2026-10-20
- DelawareNot found in the instruments we track
- FloridaNot found in the instruments we track
- GeorgiaNot found in the instruments we track
- HawaiiNot found in the instruments we track
- IdahoNot found in the instruments we track
- IllinoisAdopted — not yet in force
Frontier developers must report critical safety incidents, alongside broader catastrophic-risk safety duties.
- Illinois AI Safety Act, Monitoring and Evaluation — Act, Awaiting Entry (from 2027-01-01)
A possible change to a cited instrument is not yet verified.
Status verified 2026-09-09
- IndianaNot found in the instruments we track
- IowaNot found in the instruments we track
- KansasNot found in the instruments we track
- KentuckyNot found in the instruments we track
- LouisianaNot found in the instruments we track
- MaineNot found in the instruments we track
- MarylandNot found in the instruments we track
- MassachusettsNot found in the instruments we track
- MichiganNot found in the instruments we track
- MinnesotaNot found in the instruments we track
- MississippiNot found in the instruments we track
- MontanaNot found in the instruments we track
- NebraskaNot found in the instruments we track
- NevadaNot found in the instruments we track
- New HampshireNot found in the instruments we track
- New JerseyNot found in the instruments we track
- New MexicoNot found in the instruments we track
- New YorkAdopted — not yet in force
Large frontier developers must implement and publish safety frameworks addressing catastrophic risks, while frontier developers must report critical incidents.
- New York RAISE Act Amendments, § 1421(1) — Act, Awaiting Entry (from 2027-01-01)
- New York RAISE Act for AI Safety, § 1421(1) — Act, Awaiting Entry (from 2027-01-01)
Status verified 2026-09-08
- North CarolinaNot found in the instruments we track
- North DakotaNot found in the instruments we track
- OhioNot found in the instruments we track
- OklahomaNot found in the instruments we track
- OregonPartly — regional, framework or general (non-AI) law
The State must seek independently safety-reviewed frontier AI models for procurement, contracting, or use.
- Oregon Responsible AI Procurement Order, Overview — Decree, In Force
Status verified 2026-10-20
- PennsylvaniaNot found in the instruments we track
- Puerto RicoNot found in the instruments we track
- Rhode IslandNot found in the instruments we track
- South CarolinaNot found in the instruments we track
- South DakotaNot found in the instruments we track
- TennesseeNot found in the instruments we track
- TexasNot found in the instruments we track
- UtahNot found in the instruments we track
- VermontNot found in the instruments we track
- VirginiaNot found in the instruments we track
- WashingtonNot found in the instruments we track
- West VirginiaNot found in the instruments we track
- WisconsinNot found in the instruments we track
- WyomingNot found in the instruments we track
Proposed (2)
Brazil
ProposedDevelopers of general-purpose generative AI with systemic risk must document tests and analyses before commercial release.
- Brazil - National AI Regulation (2.338/2023), Art. 30, II — Bill, Under Review
Status verified 2026-09-08
United Kingdom
ProposedThe Bill bans ASI systems because of severe national-security and existential risks from superintelligence.
- UK Superintelligence Ban Bill, Overview — Bill, Proposed
- United Kingdom - AI Safety Cooperation — Policy, Adopted (from 2023-11-01)
Status verified 2026-10-08
Guidance or policy only (4)
Canada
Guidance or policy onlyIt encourages additional mitigations for widely available advanced generative AI systems based on anticipated impact and exposure.
- Canada - Responsible AI Development, Implementation Framework — Guideline, In Force
Status verified 2026-09-07
0 of 3 regions address this
- British ColumbiaNot found in the instruments we track
- OntarioNot found in the instruments we track
- QuebecNot found in the instruments we track
China
Guidance or policy onlyDevelopers are recommended to test regularly for technological loss-of-control risks.
- China AI Safety Governance Framework 3.0, Section 4.5.1 — Guideline, Adopted
Status verified 2026-10-20
0 of 4 regions address this
- BeijingNot found in the instruments we track
- ShanghaiNot found in the instruments we track
- ShenzhenNot found in the instruments we track
- ZhejiangNot found in the instruments we track
Italy
Guidance or policy onlyPublic administrations must verify that systemic-risk general-purpose AI providers fulfilled AI Act obligations, including independent testing.
- Italy - AI Adoption Guidelines — Guideline, Draft
Status verified 2026-09-08
As an EU member state, the EU AI Act also applies here.
0 of 2 regions address this
- Emilia-RomagnaNot found in the instruments we track
- LombardyNot found in the instruments we track
Japan
Guidance or policy onlyAdvanced-system developers are encouraged to refer to the Hiroshima Process code of conduct.
- Japan Business AI Guidelines 1.2, Part 3 — Guideline, Adopted
Not found in the instruments we track (77)
Argentina
Not found in the instruments we track0 of 5 regions address this
- Buenos Aires ProvinceNot found in the instruments we track
- City of Buenos AiresNot found in the instruments we track
- JujuyNot found in the instruments we track
- Río NegroNot found in the instruments we track
- San JuanNot found in the instruments we track
Australia
Not found in the instruments we track0 of 7 regions address this
- Australian Capital TerritoryNot found in the instruments we track
- New South WalesNot found in the instruments we track
- Northern TerritoryNot found in the instruments we track
- QueenslandNot found in the instruments we track
- TasmaniaNot found in the instruments we track
- VictoriaNot found in the instruments we track
- Western AustraliaNot found in the instruments we track
Austria
Not found in the instruments we trackAs an EU member state, the EU AI Act also applies here.
Bahrain
Not found in the instruments we trackBangladesh
Not found in the instruments we trackBelgium
Not found in the instruments we trackAs an EU member state, the EU AI Act also applies here.
Bulgaria
Not found in the instruments we trackAs an EU member state, the EU AI Act also applies here.
Chile
Not found in the instruments we trackColombia
Not found in the instruments we trackCosta Rica
Not found in the instruments we trackCôte d'Ivoire
Not found in the instruments we trackCroatia
Not found in the instruments we trackAs an EU member state, the EU AI Act also applies here.
Cyprus
Not found in the instruments we trackAs an EU member state, the EU AI Act also applies here.
Czech Republic
Not found in the instruments we trackAs an EU member state, the EU AI Act also applies here.
Denmark
Not found in the instruments we trackAs an EU member state, the EU AI Act also applies here.
Ecuador
Not found in the instruments we trackEgypt
Not found in the instruments we trackEstonia
Not found in the instruments we trackAs an EU member state, the EU AI Act also applies here.
Ethiopia
Not found in the instruments we trackFinland
Not found in the instruments we trackAs an EU member state, the EU AI Act also applies here.
0 of 1 regions address this
- HelsinkiNot found in the instruments we track
France
Not found in the instruments we trackAs an EU member state, the EU AI Act also applies here.
Germany
Not found in the instruments we trackAs an EU member state, the EU AI Act also applies here.
0 of 3 regions address this
- Baden-WürttembergNot found in the instruments we track
- BavariaNot found in the instruments we track
- North Rhine-WestphaliaNot found in the instruments we track
Ghana
Not found in the instruments we trackGreece
Not found in the instruments we trackAs an EU member state, the EU AI Act also applies here.
Hong Kong
Not found in the instruments we trackHungary
Not found in the instruments we trackAs an EU member state, the EU AI Act also applies here.
Iceland
Not found in the instruments we trackIndia
Not found in the instruments we track0 of 4 regions address this
- KarnatakaNot found in the instruments we track
- MaharashtraNot found in the instruments we track
- Tamil NaduNot found in the instruments we track
- TelanganaNot found in the instruments we track
Indonesia
Not found in the instruments we trackIran
Not found in the instruments we trackIraq
Not found in the instruments we trackIreland
Not found in the instruments we trackAs an EU member state, the EU AI Act also applies here.
Israel
Not found in the instruments we trackKazakhstan
Not found in the instruments we trackKenya
Not found in the instruments we trackLatvia
Not found in the instruments we trackAs an EU member state, the EU AI Act also applies here.
Lebanon
Not found in the instruments we trackLithuania
Not found in the instruments we trackAs an EU member state, the EU AI Act also applies here.
Luxembourg
Not found in the instruments we trackAs an EU member state, the EU AI Act also applies here.
Malaysia
Not found in the instruments we trackMaldives
Not found in the instruments we trackMalta
Not found in the instruments we trackAs an EU member state, the EU AI Act also applies here.
Mexico
Not found in the instruments we trackMorocco
Not found in the instruments we track0 of 1 regions address this
- Casablanca-SettatNot found in the instruments we track
Netherlands
Not found in the instruments we trackAs an EU member state, the EU AI Act also applies here.
0 of 1 regions address this
- AmsterdamNot found in the instruments we track
New Zealand
Not found in the instruments we trackNigeria
Not found in the instruments we trackNorway
Not found in the instruments we trackPakistan
Not found in the instruments we trackPanama
Not found in the instruments we trackPeru
Not found in the instruments we trackPhilippines
Not found in the instruments we trackPoland
Not found in the instruments we trackAs an EU member state, the EU AI Act also applies here.
Portugal
Not found in the instruments we trackAs an EU member state, the EU AI Act also applies here.
Qatar
Not found in the instruments we track0 of 1 regions address this
- DohaNot found in the instruments we track
Romania
Not found in the instruments we trackAs an EU member state, the EU AI Act also applies here.
Russia
Not found in the instruments we track0 of 1 regions address this
- MoscowNot found in the instruments we track
Rwanda
Not found in the instruments we trackSaudi Arabia
Not found in the instruments we trackSenegal
Not found in the instruments we trackSerbia
Not found in the instruments we trackSingapore
Not found in the instruments we trackSlovakia
Not found in the instruments we trackAs an EU member state, the EU AI Act also applies here.
Slovenia
Not found in the instruments we trackAs an EU member state, the EU AI Act also applies here.
South Africa
Not found in the instruments we trackSpain
Not found in the instruments we trackAs an EU member state, the EU AI Act also applies here.
0 of 3 regions address this
- AsturiasNot found in the instruments we track
- CataloniaNot found in the instruments we track
- GaliciaNot found in the instruments we track
Sweden
Not found in the instruments we trackAs an EU member state, the EU AI Act also applies here.
Switzerland
Not found in the instruments we trackTaiwan
Not found in the instruments we track0 of 1 regions address this
- KaohsiungNot found in the instruments we track
Thailand
Not found in the instruments we trackTunisia
Not found in the instruments we trackTurkey
Not found in the instruments we trackUkraine
Not found in the instruments we trackUnited Arab Emirates
Not found in the instruments we track0 of 1 regions address this
- DubaiNot found in the instruments we track
Uruguay
Not found in the instruments we trackUzbekistan
Not found in the instruments we trackVietnam
Not found in the instruments we track0 of 1 regions address this
- Ho Chi Minh CityNot found in the instruments we track
International organisations (23)
African Union
Not found in the instruments we trackAPEC
Not found in the instruments we trackASEAN
Not found in the instruments we trackBRICS
Not found in the instruments we trackCouncil of Europe
Not found in the instruments we trackEuropean Union
Yes — binding law in forceProviders of general-purpose AI models with systemic risk must assess and mitigate systemic risks.
- EU AI Act, Article 55(1) — Regulation, In Force (Amended)
- European Union - General-Purpose AI Guidelines, Article 55 AI Act — Guideline, In Force
- EU General-Purpose AI Code of Practice — Guideline, In Force
A possible change to a cited instrument is not yet verified.
Status verified 2026-09-11
G7
Guidance or policy onlyOrganizations developing advanced AI must identify, evaluate, and mitigate lifecycle risks, including through pre-deployment testing.
- G7 Hiroshima AI Process, Principle 1 — Policy, Adopted (from 2023-12-01)
Status verified 2026-09-09
GPAI
Not found in the instruments we trackIAEA
Not found in the instruments we trackICAO
Not found in the instruments we trackIEEE
Not found in the instruments we trackILO
Not found in the instruments we trackInterpol
Not found in the instruments we trackISO
Not found in the instruments we trackITU
Not found in the instruments we trackNATO
Not found in the instruments we trackOAS
Not found in the instruments we trackOECD
Not found in the instruments we trackUNESCO
Not found in the instruments we trackUnited Nations
Not found in the instruments we trackWHO
Not found in the instruments we trackWorld Bank
Not found in the instruments we trackWorld Economic Forum
Not found in the instruments we track
What the AI companies have committed to
At the AI Seoul Summit in May 2024, 20 AI companies promised to publish a safety framework for severe risks from their most capable models. 12 have published one, 2 published something narrower and 6 have published nothing. These are voluntary pledges, not law. Checked 2026-09-30.
- Amazon: Amazon's Frontier Model Safety Framework
- Anthropic: Responsible Scaling Policy, Version 3.4
- Cohere: The Cohere Secure AI Frontier Model Framework V1.0
- G42: G42's Frontier AI Safety Framework
- Google: Frontier Safety Framework, Version 3.1
- Magic: AGI Readiness Policy
- Meta: Meta Superintelligence Scaling Framework, Version 2.1
- Microsoft: Frontier Governance Framework
- Naver: NAVER ASF 2.0: AI Safety Framework 2.0
- NVIDIA: Frontier AI Risk Assessment
- OpenAI: Preparedness Framework, Version 2
- xAI: xAI Frontier Artificial Intelligence Framework
Who published what, and who did not·How company positions compare with the law
Research on this topic
- What If Automating AI R&D Triggers an Intelligence Explosion? (2026)
- Embedded Assessments for Frontier AI (2026)
- Thematic Brief on AI Agents, Misalignment and the Risk of Losing Human Control (2026)
- Improving Frontier AI Incident Reporting Regimes (2026)
- Harmonizing AI Safety Thresholds (2026)
- The 2026 Singapore Consensus on Global AI Safety Research Priorities (2026)
- Preliminary Report of the Independent International Scientific Panel on AI: Evidence-based assessment of opportunities, risks and impacts of AI (2026)
- Verifying Restrictions on Frontier AI Research (2026)