Skip to content
ContentLora

    Tip: press / anywhere to search.

    Explainer

    AI regulation around the world: EU, US, UK, China and South Korea

    Governments regulate AI in very different ways. The EU has a risk-based AI Act that became generally applicable on 2 August 2026, and South Korea's AI Basic Act took effect on 22 January 2026.[1][2][3] China regulates generative AI through targeted rules in force since 2023.[4] In 2025 the US federal government set AI policy through executive orders, and the UK's 2023 white paper chose existing regulators over a new law.[5][6][7]

    Editor reviewedStrict sourcingUpdated Explainer33 claims · 100% primaryTech policy, Artificial intelligence
    Reading level: 3 passages differ; the rest is the same for everyone.

    This explainer compares how five jurisdictions approach AI regulation. Laws’ statuses are given as of 7 November 2026, based on the official texts and other sources cited. It describes laws and official texts; it does not take a side on which approach is better, and it is not legal advice. For compliance decisions, read the official text or ask a lawyer. Chinese texts are cited from the official CAC pages, with an unofficial English translation alongside.

    There is no single global AI law. The European Union has a broad, risk-based statute.[1] China has a set of targeted rules for specific technologies.[4][8] South Korea has a framework act that combines support for industry with baseline duties.[3][9] The United States federal government has acted through executive orders, and the United Kingdom’s 2023 white paper chose existing regulators over a new law.[5][7]

    European Union: the AI Act

    The EU AI Act sorts AI by risk. Some uses are banned, high-risk uses face strict rules, chatbots and deep fakes must be disclosed, and most AI has no special rules.[1][10][11][12] It entered into force on 1 August 2024 and became generally applicable on 2 August 2026.[13][2]

    Regulation (EU) 2024/1689 was adopted on 13 June 2024.[14] The first eight prohibitions applied from 2 February 2025, and the rules for general-purpose AI models from 2 August 2025.[15][16] The 2026 Digital Omnibus moved the high-risk requirements to 2 December 2027 for Annex III systems and 2 August 2028 for products under Annex I.[17] Fines for prohibited practices reach 35 million euros or 7% of worldwide turnover, whichever is higher.[18] The Act reaches providers wherever they are established if they place AI on the EU market.[19]

    For the full structure and dates, read the EU AI Act explained and the EU AI Act timeline.

    United States: executive orders and state laws

    In 2025 US federal AI policy was set through presidential orders. In January 2025 an order revoked AI policies it called barriers to innovation and ordered a review of actions taken under the previous administration’s main AI order.[5] In December 2025 another order set up a federal task force to challenge state AI laws that conflict with federal policy.[6]

    Executive Order 14179, signed on 23 January 2025, revoked AI policies the administration called barriers to innovation and ordered a review of actions taken under Executive Order 14110 of 30 October 2023, which it described as revoked.[5] The July 2025 AI Action Plan backed open-source and open-weight models as important for startups, research and geostrategic value.[20] Executive Order 14365 of 11 December 2025 directed the Attorney General to create an AI Litigation Task Force to challenge state AI laws inconsistent with its policy.[6] It also asked advisers to prepare legislation for a uniform federal framework that preempts conflicting state laws, while not proposing to preempt state child-safety laws.[21]

    States have passed their own laws. Texas HB 149 took effect on 1 January 2026.[22] New York’s RAISE Act, signed on 19 December 2025, requires large AI developers to publish safety information and report incidents within 72 hours, according to the governor.[23] Colorado revised its 2024 AI law in 2026.[24] Our US state AI laws tracker follows these laws and the court fights over them.

    United Kingdom: existing regulators

    The UK government’s March 2023 white paper said it did not initially intend to introduce new legislation. It set five cross-sector principles: safety, security and robustness; transparency and explainability; fairness; accountability and governance; and contestability and redress. Existing regulators were to apply them on a non-statutory basis.[7] In January 2025 the government published the AI Opportunities Action Plan, which recommends AI Growth Zones to speed up data-centre building.[25] This page does not cite an official source on whether the UK has legislated on AI since then.

    The uk-ai-security-institute is a research organisation within the Department for Science, Innovation and Technology.[26] It was renamed from the AI Safety Institute in February 2025 to focus on security risks such as cyberattacks and chemical and biological weapons.[27]

    China: rules for specific technologies

    China has written rules for particular kinds of AI rather than one broad law. Its rules for generative AI services, such as chatbots, took effect on 15 August 2023.[4] Since 1 September 2025, AI-generated content must carry labels, both visible ones and hidden ones such as watermarks.[8]

    The Interim Measures for the Management of Generative AI Services were issued by the Cyberspace Administration of China with six other bodies and apply to generative AI services offered to the public in mainland China.[4] Article 4 requires services to uphold the Core Socialist Values and bars content prohibited by law, such as content inciting subversion.[28] Under Article 17, services with public-opinion properties or the capacity for social mobilisation must undergo security assessments and file their algorithms.[29] The Measures for Labeling AI-Generated Synthetic Content, issued by the CAC, MIIT, the Ministry of Public Security and the NRTA, took effect on 1 September 2025 and require explicit and implicit labels.[8]

    South Korea: the AI Basic Act

    This section follows the US International Trade Administration’s summary of the law. South Korea’s AI Basic Act and its Enforcement Decree took effect on 22 January 2026, creating a framework that combines support for the AI industry with baseline obligations for trust, safety, transparency and user protection.[3] Providers using high-impact AI or generative AI must tell users in advance that AI is being used, and may also need to disclose that AI-generated output was made by AI.[9] Large-scale advanced AI systems may face extra risk-management duties if they meet criteria such as training compute above 10^26 operations.[30]

    A foreign AI firm with no address or office in Korea may need to appoint a domestic representative if it meets set thresholds, such as prior-year revenue of at least 1 trillion won.[31] The science ministry is running a grace period of at least one year in 2026, during which fines are generally deferred.[32]

    How the approaches compare

    JurisdictionMain instrumentStatus as of 7 November 2026Claim
    EUAI Act (regulation)Generally applicable 2 August 2026; high-risk rules from December 2027[2][17]
    US (federal)Executive ordersEO 14179 signed January 2025; EO 14365 signed December 2025[5][6]
    UKNon-statutory principles for existing regulatorsSet out in a March 2023 white paper[7]
    ChinaGenerative AI measures; labelling measuresIn force since 15 August 2023 and 1 September 2025[4][8]
    South KoreaAI Basic ActIn force since 22 January 2026[3]

    Labelling or disclosure of AI-generated content is one area where several jurisdictions overlap: the EU and China require it, and Korea may require disclosure in some cases.[33][8][9] For how these rules touch everyday chatbot use, read how AI chatbots use your data.

    UK AI Security Institute (AISI)

    The UK AI Security Institute tests frontier AI models for national-security risks. Its history, tools and key findings.

    Questions readers ask

    Which country has the most comprehensive AI law?

    The European Commission describes the EU AI Act as the first comprehensive legal framework on AI worldwide. South Korea's AI Basic Act, in force since January 22, 2026, combines support for the AI industry with baseline obligations, according to the US International Trade Administration.[1][3]

    Does the US have a federal AI law?

    US federal policy has been set mainly by executive orders. Executive Order 14179 of January 2025 revoked earlier AI policies, and Executive Order 14365 of December 2025 targets state AI laws and asks for a federal framework that would preempt conflicting ones.[5][6][21]

    How does the UK regulate AI?

    The government's 2023 white paper set five principles for existing regulators to apply on a non-statutory basis, and said the government did not initially intend to introduce new legislation.[7]

    What are China's rules for AI chatbots?

    China's Interim Measures for generative AI services took effect on August 15, 2023. They require services to uphold the Core Socialist Values, and services that can shape public opinion must pass security assessments and file their algorithms. Labelling rules for AI-generated content followed on September 1, 2025.[4][28][29][8]

    Do AI laws apply to foreign companies?

    Often yes. The EU AI Act covers providers placing AI on the EU market wherever they are based. In Korea, a foreign AI firm with no office there may need to appoint a local representative if it meets thresholds such as prior-year revenue of at least 1 trillion won, according to the US International Trade Administration.[19][31]

    Are AI-generated images required to be labelled?

    In several places. The EU AI Act requires machine-readable marking of AI-generated content, China's labelling measures require explicit and implicit labels, and in Korea providers may need to disclose that output was made by AI.[33][8][9]

    Sources

    · 33 claims from 16 publishers

    Each numbered claim is a statement we checked against the sources listed with it. Tap a number in the text to see its sources in place.

    1. [1]

      The European Commission describes the AI Act as the first comprehensive legal framework on AI worldwide, setting risk-based rules for AI developers and deployers. confirmedas of 2026-11-07

    2. [2]

      The AI Act's transparency rules took effect in August 2026, when the Act became generally applicable on 2 August 2026. confirmedas of 2026-08-02

    3. [3]

      According to the US International Trade Administration, South Korea's AI Basic Act and its Enforcement Decree took effect on 22 January 2026, creating a national framework that combines support for the AI industry with baseline obligations for trust, safety, transparency and user protection. confirmedas of 2026-11-07

    4. [4]

      China's Interim Measures for the Management of Generative Artificial Intelligence Services, issued by the Cyberspace Administration of China with six other national bodies, took effect on 15 August 2023 and apply to generative AI services offered to the public in mainland China. confirmedas of 2026-11-07

    5. [5]

      Executive Order 14179, Removing Barriers to American Leadership in Artificial Intelligence, signed on 23 January 2025, revoked AI policies the administration described as barriers to innovation and ordered a review of actions taken under Executive Order 14110 of 30 October 2023, which it described as revoked. confirmedas of 2025-01-23

    6. [6]

      Executive Order 14365, Ensuring a National Policy Framework for Artificial Intelligence, signed on 11 December 2025, directed the Attorney General to set up within 30 days an AI Litigation Task Force whose sole job is to challenge state AI laws inconsistent with the order's policy. confirmedas of 2025-12-11

    7. [7]

      The UK government's March 2023 AI regulation white paper set out five cross-sector principles - safety, security and robustness; appropriate transparency and explainability; fairness; accountability and governance; and contestability and redress - to be issued on a non-statutory basis and implemented by existing regulators, and said the government did not initially intend to introduce new legislation. confirmedas of 2023-03-29

    8. [8]

      China's Measures for Labeling AI-Generated Synthetic Content, issued by the Cyberspace Administration of China, the Ministry of Industry and Information Technology, the Ministry of Public Security and the National Radio and Television Administration, took effect on 1 September 2025 and require both explicit labels visible to users and implicit labels such as metadata or watermarks embedded in files. confirmedas of 2026-11-07

    9. [9]

      According to the US International Trade Administration, providers using high-impact AI or generative AI under Korea's AI Basic Act must notify users in advance that AI is being used, and providers may also need to disclose that AI-generated output was generated by AI. confirmedas of 2026-11-07

    10. [10]

      The AI Act prohibits nine practices, including harmful AI-based manipulation and deception, social scoring, untargeted scraping of facial images to build recognition databases, emotion recognition in workplaces and schools, and real-time remote biometric identification for law enforcement in public spaces; the ninth, on AI that generates non-consensual intimate content or child sexual abuse material, was added by the Omnibus. confirmedas of 2026-11-07

    11. [11]

      Under the AI Act's transparency rules, people must be made aware when they are interacting with an AI system such as a chatbot, and certain AI-generated content, namely deep fakes and text published to inform the public on matters of public interest, must be clearly labelled. confirmedas of 2026-11-07

    12. [12]

      The AI Act sets no rules for minimal- or no-risk AI, which the Commission says covers the vast majority of AI systems used in the EU, such as AI-enabled video games and spam filters. confirmedas of 2026-11-07

    13. [13]

      The AI Act entered into force on 1 August 2024. confirmedas of 2024-08-01

    14. [14]

      The AI Act is Regulation (EU) 2024/1689 of the European Parliament and of the Council of 13 June 2024, published in the Official Journal on 12 July 2024. confirmedas of 2024-07-12

    15. [15]

      The first eight prohibitions, together with the AI literacy obligations, applied from 2 February 2025. confirmedas of 2025-02-02

    16. [16]

      The AI Act's rules for general-purpose AI models and its governance rules applied from 2 August 2025. confirmedas of 2025-08-02

    17. [17]

      The Omnibus moved the high-risk requirements to 2 December 2027 for systems listed in Annex III and to 2 August 2028 for systems covered by the Annex I product laws. confirmedas of 2026-07-27

    18. [18]

      Breaching the prohibitions in Article 5 can bring fines of up to 35 million euros or 7% of a company's total worldwide annual turnover, whichever is higher. confirmedas of 2024-07-12

    19. [19]

      The AI Act applies to providers that place AI systems or general-purpose AI models on the EU market whether they are established in the EU or in a third country, and to deployers of AI systems established or located in the EU. confirmedas of 2024-07-12

    20. [20]

      The White House's July 2025 AI Action Plan said open-source and open-weight models help startups and organisations with sensitive data, are essential for academic research, and have geostrategic value, and that America needs leading open models. confirmedas of 2025-07-23

    21. [21]

      The order asked White House advisers to prepare a legislative recommendation for a uniform federal AI framework that preempts conflicting state laws, while not proposing to preempt otherwise lawful state laws on child safety and certain other areas. confirmedas of 2025-12-11

    22. [22]

      Texas HB 149 regulates the use of AI systems in the state with civil penalties; it was signed on 22 June 2025 and took effect on 1 January 2026. confirmedas of 2026-01-01

    23. [23]

      On 19 December 2025 New York's governor signed the Responsible AI Safety and Education (RAISE) Act, S6953B; she said the agreed chapter amendments require large AI developers to publish information about their safety protocols and report incidents to the state within 72 hours. confirmedas of 2025-12-19

    24. [24]

      Colorado's SB26-189, signed by the governor on 14 May 2026, revised the 2024 law, repealing and re-enacting it with new requirements on automated decision-making technology (ADMT) that influences consequential decisions, beginning in 2027. confirmedas of 2026-05-14

    25. [25]

      The UK government published the AI Opportunities Action Plan on 13 January 2025; it recommends measures such as AI Growth Zones to speed up the build-out of AI data centres. confirmedas of 2025-01-13

      • AI Opportunities Action Plan · GOV.UK (Department for Science, Innovation and Technology) · Primary source · 2025-01-13 (retrieved 2026-11-07)
      • AI Opportunities Action Plan · GOV.UK (Department for Science, Innovation and Technology) · Primary source · 2025-01-13 (retrieved 2026-11-07)
    26. [26]

      The UK AI Security Institute (AISI) is a research organisation within the UK Department for Science, Innovation and Technology, with more than 100 technical staff and £66 million in funding per financial year. confirmedas of 2026-10-10

    27. [27]

      On 14 February 2025 the UK AI Safety Institute was renamed the AI Security Institute, with a sharper focus on chemical and biological weapons, cyberattacks and criminal misuse, and no longer focusing on bias or freedom of speech. confirmedas of 2025-02-14

    28. [28]

      Article 4 of China's generative AI measures requires services to uphold the Core Socialist Values and not generate content prohibited by law, such as content inciting subversion of state power. confirmedas of 2026-11-07

    29. [29]

      Under Article 17 of China's generative AI measures, providers of generative AI services with public-opinion properties or the capacity for social mobilisation must carry out security assessments and file their algorithms. confirmedas of 2026-11-07

    30. [30]

      According to the US International Trade Administration, large-scale advanced AI systems may face additional risk identification, assessment, mitigation and management obligations under Korea's framework if they meet criteria such as training compute above 10^26 floating-point operations. confirmedas of 2026-11-07

    31. [31]

      According to the US International Trade Administration, a foreign AI business operator without an address or business office in Korea may need to appoint a domestic representative under the Enforcement Decree if it meets set thresholds, such as prior-year total revenue of at least 1 trillion won. confirmedas of 2026-11-07

    32. [32]

      According to the US International Trade Administration, Korea's science ministry (MSIT) is operating a grace period of at least one year in 2026 during which administrative fines under the AI Basic Act are generally deferred. confirmedas of 2026-11-07

    33. [33]

      Article 50(2) requires providers of AI systems that generate synthetic audio, image, video or text to mark the outputs in a machine-readable format so they are detectable as artificially generated or manipulated. confirmedas of 2024-07-12

    Revision history (1)
    1. Page created.

    Created Nov 7, 2026. Last reviewed by an editor on Nov 8, 2026. Next scheduled review: Feb 7, 2027.

    Cite this page

    "AI regulation around the world: EU, US, UK, China and South Korea." ContentLora, updated Nov 7, 2026. https://contentlora.com/explain/ai-regulation-around-the-world-explained

    Spotted an error? Suggest a correction or emailcorrections@contentlora.com.