Global AI Regulatory Tracker

New framework and APDP take effect; Law 19,628 remains operative until then.

Art. 50(2) machine-readable marking for AI systems placed on the market before 2 August 2026 (new Art. 111(4) of Regulation (EU) 2024/1689: four months, to 2 December 2026), and the new prohibitions on non-consensual intimate imagery and CSAM generation apply.

Member States must transpose into national law.

Schedule 1 Part 15 of Act No. 128 of 2024 commences the day after the end of 24 months beginning on Royal Assent (s. 2(1) table item 7).

General operative date; disclosure filings and daily penalties.

§ 3.2.3(v): a midterm review of the Continental AI Strategy in 2027 to refine indicators and improve implementation; § 3.2.1 makes Phase II conditional on that review. Year only in the source; recorded at the start of the year.

Sch 1 item 45 of the Interactive Gambling Amendment (Gambling Reform) Act 2026 (No. 72, 2026), recorded on the Federal Register of Legislation as an uncommenced amendment to this Act.

C.R.S. § 6-1-1704(4)(b) and § 6-1-1705(3): rules on post-adverse-outcome disclosures and on correction and human review.

S.B. 26-189 § 5: takes effect January 1, 2027 and applies to consequential decisions made on or after that date.

Pre-use notice / ADMT Article 11 operational compliance deadline for significant decisions (11 CCR §7200).

Large online platforms and GenAI hosting platforms.

Secs. 5–6 AI companion disclosure and minor-safety duties take effect.

Sec. 2(c) internal anonymous safety-reporting process required for large frontier developers (group annual gross revenue over $500 million).

Secs. 5-6 AI companion disclosure and minor-safety duties take effect. Current treatment: connecticut-cart-act.

Sec. 2(c) internal anonymous safety-reporting process required for large frontier developers (group annual gross revenue over $500 million). Current treatment: connecticut-cart-act.

85 Del. Laws c. 463 rewrites 6 Del. C. §§ 12D-102, -103, -104, -106, -108 and -109. § 12D-103(a) as amended applies to controllers of 10,000 consumers (5,000 with more than 20% of gross revenue from the sale of personal data) and to third parties who acquire personal data from a controller.

Section 10 of the act: "Sections 3 and 7 of this act take effect January 1, 2027." Section 3 is a reenactment of RCW 48.43.830 reconciling a double amendment with 2025 c 227 s 8, and section 7 does the same for RCW 48.43.535. The artificial-intelligence provisions are identical in both versions of RCW 48.43.830, so no duty changes on this date.

Section 9 of the act: "Sections 2 and 6 of this act expire January 1, 2027." The expiry is the other half of the double-amendment reconciliation and is simultaneous with the section 10 commencement, so no provision lapses without a replacement.

Iowa Code § 514F.8A (peer review of prior authorization denials and downgrades) and § 514F.8C (claim audits).

Section 2 of the enrolled Act: substantive Chapter 20-B duties, including the § 1780.4(E) data protection assessment, commence for processing activities as of this date; not retroactive.

Ontario municipal and local-board privacy law — access, collection, use, and disclosure rules for municipal personal information.

RAISE Act obligations on covered AI systems and frontier model developers take effect.

§ 507-H:6, I(e-1) (ban on sale of a child's personal data), added by 2026 N.H. Laws ch. 168, takes effect January 1, 2027; carried by the row nhdpa:provision:child-data-sale-ban as upcoming.

O.C.G.A. Title 33 Chapter 46 — utilization review AI limits.

Section 22 of the enrolled Act: substantive Title 75A duties, including the § 309 data protection assessment, commence for processing activities on or after this date; not retroactive.

Printed on the chaptered act; the act has no effective-date clause, so ORS 171.022 applies (January 1 of the year after passage). Both sections commence together; nothing is phased.

Pt 2 Div 6 (notifiable information breaches), Div 10 Subdiv 4, ss 134-136 and Pt 3 Div 6 commence (Commencement Proclamation (No. 2) 2026, SL 2026/203, cl. 2, published 1 October 2026).

Title 6 Chapter 63 — crisis protocols and three-hour AI-not-human notifications.

Order under s. 1(3) dated 13 July 2026, Gazette Extraordinary No. 2498/16 of 22 July 2026. Part II, Part IV and Part VII are not named in it.

Counterfeit intimate images, generation services, and covered-platform provenance duties.

Sec. 9 of the act: "This act takes effect January 1, 2027." All nine sections commence together; nothing is phased.

Sec. 4(b) of the act: the Council must submit a written report to the General Assembly recommending statutory changes, including neurological-rights and neurotechnology protections, generative-AI guidance for regulated professions, and AI/augmented-intelligence regulation in health-insurance utilization review.

Art. 101(2): arts. 29(3), 31, 32, 35, 36, 64, 65 and 67(2), (3) and (5) enter into force 2 years after publication — the impact assessment (art. 31), prior consultation (art. 32) and the duty to inform the data subject of a breach (art. 29(3)) among them.

Sec. 9 of the act: "This act takes effect February 1, 2027." Both chapters the act creates commence together; nothing is phased.

§ 14-4714 discretionary 60-day cure opportunity applies only to alleged violations occurring on or before this date.

Section 12 of the enrolled Act: "This act shall become effective on May 1, 2027." Substantive duties are not retroactive.

OSFI's principles-based guideline on enterprise-wide model risk management for all Federally Regulated Financial Institutions, including AI/ML models.

Main compliance obligations phase in 18 months after Rules notification; Data Protection Board established immediately.

Conversational AI disclosure, minor safeguards, crisis protocols, and parental controls.

Operators must begin annual suicide-prevention protocol reports to the Office of Suicide Prevention.

An Act amending Titles 35 (Health and Safety), 40 (Insurance) and 67 (Public Welfare) of the Pennsylvania Consolidated Statutes, providing for artificial intelligence in facilities, for artificial intelligence use by insurers and for artificial intelligence use by MA or CHIP managed care plans; imposing duties on the Department of Health, the Insurance Department and the Department of Human Services; and imposing penalties.

An act to add Part 5.6 (commencing with Section 7200) to Division 4 of the Civil Code, relating to consumer protection.

Relative to health carrier recordkeeping requirements in utilization review, including specifications regarding the use of artificial intelligence.

Artificial Intelligence Chatbots and Minors Act established; enforcement; civil penalties; individual action.

Creates the Protection from Surveillance Pricing Act.

An Act To Amend Title 14 Of The Delaware Code Relating To The Artificial Intelligence And Extended Reality In Schools Pilot Program.

Social Media & AI Safety

INSURANCE -- THE TRANSPARENCY AND ACCOUNTABILITY IN ARTIFICIAL INTELLIGENCE USE BY HEALTH INSURERS TO MANAGE COVERAGE AND CLAIMS ACT - Promotes transparency and accountability in the use of artificial intelligence by health insurers to manage coverage and claims.

Prohibits the use of electronic shelving labels, digital shelf display technology, and surveillance pricing in food retail establishments and drug retail establishments; provides injunctive relief and civil penalties.

To establish a task force to study the effects of the use of artificial intelligence on the trades industry and require the Commissioner of Public Health to study the feasibility of establishing a certified nursing assistant training program.