EU AI Act to require machine-readable marking on older generative AI

1. 2 Dec 2026 EU AI Act to require machine-readable marking on older generative AI Providers of systems that make synthetic audio, image, video or text, placed on the market before 2 August 2026, must mark their output as Article 50(2) requires. For: providers of generative AI systems on the market before 2 August 2026. Providers of AI systems, including general-purpose AI systems, generating synthetic audio, image, video or text content, that have been placed on the market before 2 August 2026 shall take the necessary steps in order to comply with Article 50(2) by 2 December 2026. — Regulation (EU) 2026/1744 (Digital Omnibus on AI), Official Journal L, 24.7.2026
2. 2 Dec 2026 EU AI Act to ban AI made for non-consensual sexual imagery Two bans that the Digital Omnibus on AI added to Article 5 apply. They cover AI systems that make realistic sexual images, video or audio of an identifiable person without consent, and AI systems that make child sexual abuse material. A system is banned when that is its purpose, or when it can readily do so and lacks adequate safeguards. A deployer must not use a system for that purpose. For: providers and deployers. Chapters I and II shall apply from 2 February 2025, with the exception of Article 5(1), first subparagraph, points (ba) and (bb), and Article 5(1a) and (1b) which shall apply from 2 December 2026 — Regulation (EU) 2026/1744 (Digital Omnibus on AI), Official Journal L, 24.7.2026
3. 1 Jan 2027 California’s AI Transparency Act to bar hosting of AI systems without disclosures A hosting platform must not knowingly offer a generative AI system that does not place the disclosures the Act requires. For: GenAI system hosting platforms. 22757.3.2. (a) A GenAI system hosting platform shall not knowingly make available a GenAI system that does not place disclosures pursuant to Section 22757.3. (b) This section shall become operative on January 1, 2027. — AB 853 (2025), Chapter 674, Statutes of 2025, as chaptered
4. 1 Jan 2027 California’s AI Transparency Act to require provenance checks on large platforms A large online platform must detect standard provenance data in content it distributes. It must show users what provenance data is available and let them inspect it. Where technically feasible, it must not knowingly remove that data. For: large online platforms. 22757.3.1. (a) A large online platform shall do all of the following: … (c) This section shall become operative on January 1, 2027. — AB 853 (2025), Chapter 674, Statutes of 2025, as chaptered
5. 1 Jan 2027 California’s frontier AI law to start yearly whistleblower reports The Attorney General starts a yearly report of anonymized, aggregated information about whistleblower reports from covered employees. For: California Attorney General. (d) (1) Beginning January 1, 2027, and annually thereafter, the Attorney General shall produce a report with anonymized and aggregated information about reports from covered employees — SB 53 (2025), Chapter 138, Statutes of 2025, as chaptered
6. 1 Jan 2027 California’s frontier AI law to start yearly reviews of its definitions The Department of Technology must review the Act's definitions of frontier model, frontier developer and large frontier developer each year, and recommend updates. For: California Department of Technology. (a) On or before January 1, 2027, and annually thereafter, the Department of Technology shall assess recent evidence and developments relevant to the purposes of this chapter — SB 53 (2025), Chapter 138, Statutes of 2025, as chaptered
7. 1 Jan 2027 California’s frontier AI law to start yearly safety-incident reports The Office of Emergency Services starts a yearly report of anonymized, aggregated information about critical safety incidents. For: California Office of Emergency Services. (g) (1) Beginning January 1, 2027, and annually thereafter, the Office of Emergency Services shall produce a report with anonymized and aggregated information about critical safety incidents — SB 53 (2025), Chapter 138, Statutes of 2025, as chaptered
8. 1 Jan 2027 Colorado’s AI law to take effect The reenacted law starts. A developer of a covered ADMT must give each deployer documentation of its intended uses, categories of training data, known limitations, and instructions for appropriate use and human review. For: developers of automated decision-making technology. The act requires the developer of an ADMT (developer) that is used to materially influence a consequential decision (covered ADMT), starting January 1, 2027, to provide a deployer of a covered ADMT (deployer) with technical documentation — SB26-189, automated decision-making technology: the General Assembly's bill page
9. 1 Jan 2027 Colorado’s AI law to require rules from the attorney general The Attorney General must adopt rules for the disclosure a deployer owes a consumer after an adverse outcome. For: Colorado Attorney General. The attorney general must adopt rules to clarify these post-adverse outcome disclosure requirements by January 1, 2027. — SB26-189, automated decision-making technology: the General Assembly's bill page
10. 1 Jan 2027 New York’s RAISE Act to take effect The Act takes effect. A large frontier developer must publish a frontier AI framework and file a disclosure statement with the office. Every frontier developer must publish a transparency report before it deploys a new or substantially modified frontier model. It must also report each critical safety incident to the office within 72 hours after it determines, or has facts to reasonably believe, that the incident occurred. For: frontier developers, and large frontier developers. § 3. This act shall take effect [on the ninetieth day after it shall have become a law] January 1, 2027. — S8828 (RAISE Act chapter amendment), Chapter 96 of the Laws of 2026: the Assembly's bill page
11. 1 Jul 2027 Utah’s AI Policy Act to be repealed The chapter is repealed. For: Utah’s Office of Artificial Intelligence Policy, and companies in its learning laboratory. (2) Title 13, Chapter 72, Artificial Intelligence Policy Act, is repealed July 1, 2027. — Utah Code Section 63I-2-213, repeal dates: Title 13
12. 2 Aug 2027 EU AI Act to bind older general-purpose AI models Providers of general-purpose AI models placed on the market before 2 August 2025 must comply with the Act. For: providers of general-purpose AI models on the market before 2 August 2025. Providers of general-purpose AI models that have been placed on the market before 2 August 2025 shall take the necessary steps in order to comply with the obligations laid down in this Regulation by 2 August 2027. — Regulation (EU) 2024/1689 (Artificial Intelligence Act), Official Journal L, 12.7.2024
13. 2 Aug 2027 EU AI Act to require an AI sandbox in each member state Each Member State must have at least one AI regulatory sandbox in operation. For: Member States. Member States shall ensure that their competent authorities establish at least one AI regulatory sandbox at national level, which shall be operational by 2 August 2027. — Regulation (EU) 2026/1744 (Digital Omnibus on AI), Official Journal L, 24.7.2026
14. 2 Dec 2027 EU AI Act to start its high-risk duties The high-risk duties of Chapter III, Sections 1 to 3, apply to Annex III systems, such as AI used in hiring, education, credit scoring and essential services. This includes risk management, data governance, technical documentation (Article 11 and Annex IV), logging and human oversight. For: providers and deployers of high-risk AI systems. Chapter III, Sections 1, 2, and 3, with the exception of Article 6(5), shall apply from: … 2 December 2027 as regards AI systems classified as high-risk pursuant to Article 6(2) and Annex III — Regulation (EU) 2026/1744 (Digital Omnibus on AI), Official Journal L, 24.7.2026
15. 1 Jan 2028 California’s AI Transparency Act to require provenance data on new capture devices For each capture device it first makes for sale in California on or after this date, a maker must embed a hidden disclosure in captured content by default, and must let users choose to include one. This applies only where technically feasible and where it complies with widely adopted standards. For: capture device manufacturers. 22757.3.3. (a) A capture device manufacturer shall, with respect to any capture device the capture device manufacturer first produced for sale in the state on or after January 1, 2028 … (c) This section shall become operative on January 1, 2028. — AB 853 (2025), Chapter 674, Statutes of 2025, as chaptered
16. 1 Jan 2028 New York’s RAISE Act to start yearly state reports The state office starts a yearly report with anonymized, aggregated information about the critical safety incidents it reviewed. For: the AI oversight office in the New York Department of Financial Services. 7. (a) Beginning January first, two thousand twenty-eight, and annual- ly thereafter, the office shall produce a report — S8828 (RAISE Act chapter amendment), Chapter 96 of the Laws of 2026: the Assembly's bill page
17. 28 Jan 2028 EU AI Act to require product notified bodies to apply for designation Notified bodies already notified under the product laws in Annex I, Section A, must apply for designation under the AI Act. For: notified bodies under the product laws in Annex I, Section A. such notified bodies which have been notified under the Union harmonisation legislation in Section A of Annex I, shall apply for designation in accordance with Section 4 of this Chapter by 28 January 2028. — Regulation (EU) 2026/1744 (Digital Omnibus on AI), Official Journal L, 24.7.2026
18. 2 Aug 2028 EU AI Act to start high-risk duties for regulated products The high-risk duties of Chapter III, Sections 1 to 3, apply to AI systems that are safety components of products under Annex I, such as medical devices and toys. For: providers and deployers of high-risk AI systems. Chapter III, Sections 1, 2, and 3, with the exception of Article 6(5), shall apply from: … 2 August 2028 as regards AI systems classified as high-risk pursuant to Article 6(1) and Annex I — Regulation (EU) 2026/1744 (Digital Omnibus on AI), Official Journal L, 24.7.2026
19. 1 Jan 2029 California’s AI Transparency Act to end its exemption for assistive technology The exemption for generative AI systems designed mainly as assistive technology ends, and the Act applies to them. From this date the hidden disclosure must also state whether a system is made mainly as assistive technology, and the penalty for false assistive-technology claims ends. For: providers of generative AI systems made as assistive technology. (b) Before January 1, 2029, this chapter does not apply to a GenAI system that is designed to primarily function as assistive technology. — SB 1000 (2026), Chapter 861, Statutes of 2026, as chaptered
20. 1 Jan 2030 Colorado’s AI law to end the notice-and-cure period before enforcement The Attorney General no longer must give 60 days' notice and a chance to cure before an enforcement action. For: developers and deployers. Before initiating an action before January 1, 2030, the attorney general must provide the developer or deployer with a 60-day notice and opportunity to cure the alleged violation, if a cure is deemed possible. — SB26-189, automated decision-making technology: the General Assembly's bill page
21. 2 Aug 2030 EU AI Act to bind public authorities’ older high-risk AI High-risk AI systems intended to be used by public authorities, and on the market before the high-risk duties applied, must comply. For: providers and deployers of high-risk AI systems used by public authorities. In any case, the providers and deployers of high-risk AI systems intended to be used by public authorities shall take the necessary steps to comply with the requirements and obligations laid down in this Regulation by 2 August 2030. — Regulation (EU) 2026/1744 (Digital Omnibus on AI), Official Journal L, 24.7.2026
22. 31 Dec 2030 EU AI Act to bind AI in large EU IT systems AI systems in the EU's large-scale IT systems in the area of freedom, security and justice (Annex X), placed on the market before 2 August 2027, must comply. For: providers and deployers of AI in the EU's large-scale IT systems. AI systems which are components of the large-scale IT systems established by the legal acts listed in Annex X that have been placed on the market or put into service before 2 August 2027 shall be brought into compliance with this Regulation by 31 December 2030. — Regulation (EU) 2024/1689 (Artificial Intelligence Act), Official Journal L, 12.7.2024

1. 1 Jan 2026 California’s AI Transparency Act to take effect · replaced, moved 213 days later, to 2 Aug 2026 The first start date of the California AI Transparency Act. AB 853 moved it to August 2, 2026. For: covered providers of generative AI systems.
2. 1 Feb 2026 Colorado’s AI law to take effect · replaced, moved 334 days later, to 1 Jan 2027 The first start date of SB24-205's duties of reasonable care against algorithmic discrimination. SB25B-004 moved it to June 30, 2026. For: developers and deployers of high-risk AI systems.
3. 30 Jun 2026 Colorado’s AI law to take effect · replaced, moved 185 days later, to 1 Jan 2027 The second start date of SB24-205's duties. SB26-189 then repealed and reenacted the law. Its developer duties start January 1, 2027. For: developers and deployers of high-risk AI systems.
4. 2 Aug 2026 EU AI Act to start its high-risk duties · replaced, moved 487 days later, to 2 Dec 2027 The first date for the high-risk duties of Chapter III, Sections 1 to 3, for Annex III systems such as hiring and credit scoring. The Digital Omnibus on AI moved it to 2 December 2027. For: providers and deployers of high-risk AI systems.
5. 2 Aug 2026 EU AI Act to require an AI sandbox in each member state · replaced, moved 365 days later, to 2 Aug 2027 The first date for each Member State to have an AI regulatory sandbox in operation. The Digital Omnibus on AI moved it to 2 August 2027. For: Member States.
6. 2 Aug 2027 EU AI Act to start high-risk duties for regulated products · replaced, moved 366 days later, to 2 Aug 2028 The first date for the high-risk duties for AI systems that are safety components of products under Annex I, such as machinery and medical devices. The Digital Omnibus on AI moved it to 2 August 2028. For: providers and deployers of high-risk AI systems.

A replaced date stays on the record, struck through, with the law that moved it. A date that a bill or proposal would set is labelled “proposed, not law” and is never in the calendar file.

• European Union · in force

  EU Artificial Intelligence Act

  EU AI Act to require machine-readable marking on older generative AI

  Regulation (EU) 2024/1689, as amended by Regulation (EU) 2026/1744 (Digital Omnibus on AI)

  The EU's horizontal AI law. It bans some AI practices, sets duties for high-risk AI systems and for general-purpose AI models, and sets transparency duties for systems that interact with people or make synthetic content. The Digital Omnibus on AI moved the high-risk dates and added two bans.

  14 dates in the law, 3 replaced

• California · in force

  California AI Transparency Act

  California’s AI Transparency Act to bar hosting of AI systems without disclosures

  SB 942, Chapter 291, Statutes of 2024, as amended by AB 853 (Chapter 674, Statutes of 2025) and SB 1000 (Chapter 861, Statutes of 2026) (Bus. & Prof. Code § 22757 et seq.)

  Providers of generative AI systems available in California must offer a free tool to check AI disclosures, and must put hidden disclosures in AI-made images, video and audio. AB 853 moved the start date. It added duties for large online platforms, hosting platforms and capture device makers. SB 1000 removed the user threshold and changed the disclosure rules again.

  6 dates in the law, 1 replaced

• California · in force

  California generative AI training data transparency law

  California’s training-data law last changed on 1 Jan 2026

  AB 2013, Chapter 817, Statutes of 2024 (Civ. Code § 3110 et seq.)

  A developer of a generative AI system or service made available to Californians must post documentation about the data used to train it.

  1 dates in the law

• California · in force

  California Transparency in Frontier Artificial Intelligence Act

  California’s frontier AI law to start yearly whistleblower reports

  SB 53, Chapter 138, Statutes of 2025 (Bus. & Prof. Code § 22757.10 et seq.)

  Large frontier developers must publish a frontier AI framework. All frontier developers must publish a transparency report for each new or substantially modified frontier model and report critical safety incidents to the state. It also protects whistleblowers.

  4 dates in the law

• Colorado · enacted; developer duties start January 1, 2027

  Colorado AI Act, repealed and reenacted as the automated decision-making technology law

  Colorado’s AI law to take effect on 1 Jan 2027

  SB24-205, delayed by SB25B-004, repealed and reenacted by SB26-189

  Developers and deployers of automated decision-making technology that materially influences consequential decisions, such as in employment, lending and housing, must give documentation and notices. After an adverse decision, a consumer can ask for human review. The start date moved twice before the 2026 law replaced the 2024 one.

  3 dates in the law, 2 replaced

• New York · enacted, not yet in effect

  New York Responsible AI Safety and Education Act (RAISE Act)

  New York’s RAISE Act to take effect on 1 Jan 2027

  S6953-B, Chapter 699 of the Laws of 2025, as amended by S8828, Chapter 96 of the Laws of 2026 (Gen. Bus. Law Art. 44-B)

  Large frontier developers must publish a frontier AI framework and file a disclosure statement with a state office. Frontier developers must publish transparency reports and report critical safety incidents to that office. The 2026 chapter amendment repealed and re-added the article and set its start date.

  4 dates in the law

• Texas · in force

  Texas Responsible Artificial Intelligence Governance Act

  Texas’s AI law last changed on 1 Sep 2026

  HB 149, 89th Legislature, Regular Session (2025)

  It bars some uses of AI, such as systems meant to encourage self-harm, harm to others or crime, and systems built with intent to discriminate unlawfully. It requires government agencies and health care providers to tell people when they deal with AI. It sets up a regulatory sandbox. The attorney general enforces it.

  2 dates in the law

• Utah · in force until its repeal date

  Utah Artificial Intelligence Policy Act

  Utah’s AI Policy Act to be repealed on 1 Jul 2027

  Utah Code Title 13, Chapter 72

  It sets up Utah's Office of Artificial Intelligence Policy, an AI learning laboratory program, and regulatory mitigation and joint interpretation agreements. A separate section of Utah law repeals the chapter on July 1, 2027.

  2 dates in the law

1. 30 Sep 2026 California’s AI Transparency Act to change its disclosure rules SB 1000 takes effect at once, as an urgency statute. It removes the user threshold from “covered provider” and renames the AI detection tool as a “disclosure verification tool”. It removes the duty to offer a visible disclosure, and adds to what the hidden disclosure must state. It delays the Act for assistive technology, and bars false claims that a system is assistive technology. For: covered providers of generative AI systems.
2. 1 Sep 2026 Texas’s AI law to require an attorney general complaint site The Attorney General must post the online form through which consumers can submit complaints under the Act. For: Texas Attorney General.
3. 2 Aug 2026 EU AI Act to start its transparency duties The rest of the Act applies, apart from the high-risk duties of Chapter III, Sections 1 to 3. This includes the transparency duties of Article 50 and the fines on general-purpose AI model providers in Article 101. For: providers and deployers.
4. 2 Aug 2026 California’s AI Transparency Act to take effect The Act takes effect. A covered provider must offer a free disclosure verification tool. Where it is technically feasible, it must also put a hidden (latent) disclosure in image, video and audio content that its generative AI system makes or changes. For: covered providers of generative AI systems (as SB 1000 amended the Act).
5. 27 Jul 2026 EU AI Act to change other EU laws Articles 102 to 110 apply. They change other EU laws, such as the vehicle and aviation rules: rules made under those laws for AI safety components must take into account the AI Act's high-risk requirements. For: the Commission and the bodies that adopt rules under those laws.
6. 27 Mar 2026 New York’s RAISE Act to be amended The Governor signs the chapter amendment, Chapter 96 of the Laws of 2026, which rewrites the article and sets the start date. For: large frontier model developers.
7. 1 Jan 2026 Texas’s AI law to take effect The Act takes effect, with its bans on some AI uses, its disclosure duties for government entities and health care, and its sandbox program. For: developers and deployers of AI systems in Texas, and government entities.
8. 1 Jan 2026 California’s frontier AI law to take effect The Act takes effect. A large frontier developer must publish a frontier AI framework. Every frontier developer must publish a transparency report before it deploys a new or substantially modified frontier model, and must report critical safety incidents to the Office of Emergency Services within 15 days. For: frontier developers.
9. 1 Jan 2026 California’s training-data law to require training-data documentation A developer must post documentation about the data used to train each generative AI system or service, released on or after January 1, 2022, that it makes available to Californians, and again before each later release or substantial change. For: developers of generative AI systems or services.
10. 19 Dec 2025 New York’s RAISE Act to become law The Governor signs the RAISE Act as Chapter 699 of the Laws of 2025. For: large frontier model developers.
11. 2 Aug 2025 EU AI Act to bind general-purpose AI model providers Chapter V applies: the duties of providers of general-purpose AI models, including technical documentation and a public summary of training content. Notified bodies, governance and penalties apply too, except the fines of Article 101. For: providers of general-purpose AI models.
12. 2 Feb 2025 EU AI Act to start its bans and AI literacy duty Chapters I and II apply: the AI literacy duty in Article 4 and the bans on prohibited AI practices in Article 5. The two bans that the Digital Omnibus on AI added apply later, from 2 December 2026. For: providers and deployers.
13. 1 Aug 2024 EU AI Act to enter into force The AI Act enters into force. Its duties apply from the later dates on this page. For: everyone.
14. 1 May 2024 Utah’s AI Policy Act to take effect The chapter takes effect. It sets up the Office of Artificial Intelligence Policy and its learning laboratory program. For: Utah’s Office of Artificial Intelligence Policy, and companies in its learning laboratory.

• Illinois: HB 3773 (AI in employment decisions) and SB 315 (AI Safety Measures Act). The General Assembly's host gave no answer to this site's fetcher, so no text is kept and no date is shown.
• Connecticut: SB 5 of 2026. The General Assembly's host gave no answer to this site's fetcher, so no text is kept and no date is shown.
• State laws on companion chatbots, provenance and deepfakes. Not in this record yet. It holds the general AI laws that bind developers and deployers first; these narrower laws are a later step.