The last 90 days: 242 AI policy changes recorded across 3 jurisdictions, of which 1 concerned binding law, 4 were rated high impact. Each is dated, links to its official source, and shows what it means in practice.
The Governor signed a package of child-safety bills that, among other things, require risk assessments before new companion-chatbot rollouts, add penalties for harm to children and restrict toys that include a companion chatbot. According to the announcement and secondary reporting, most provisions take effect on 1 January 2027 and the companion-chatbot safeguards on 1 July 2027. Recorded from secondary reporting; confirm each bill against its enrolled text.
Practical impact: Providers of companion chatbots available to minors in California should plan risk assessments and safeguards for the 2027 dates.
The Governor signed SB 813, under which the Government Operations Agency sets criteria by 1 January 2028 for designating qualified AI auditors as independent verification organisations, and AB 1405, which creates an online AI Auditor Registry by 1 January 2029, after which a covered AI audit may be offered only by a registered auditor meeting independence and integrity standards. Recorded from secondary reporting; confirm against the enrolled bills.
Practical impact: Organisations that commission AI audits for California compliance should plan to use auditors that register under AB 1405 once the registry opens.
H.R. 10044 would tax large AI companies on the value of the tokens their models process or on AI-service revenue, whichever is higher, with a rate that rises automatically as unemployment climbs, and would create a Work Protection Administration in the Department of Labor to fund jobs. Introduced by Representatives Sara Jacobs, Greg Casar and Valerie Foushee. Recorded from secondary reporting; confirm on Congress.gov.
Practical impact: No obligations; a proposal to watch in the economic-transition tracker.
The general application date in Article 113 arrived. Chatbot disclosure, synthetic-content marking and deepfake labelling under Article 50 now apply; the high-risk obligations do not, having been deferred by Regulation (EU) 2026/1744. Recorded from secondary reporting; confirm against the Official Journal text.
Practical impact: Label AI interactions and synthetic content now; high-risk conformity work runs to the deferred dates.
Regulation (EU) 2026/1744, published in the Official Journal on 24 July 2026, entered into force on 27 July 2026. It amends Article 113 of the AI Act so that the Chapter III high-risk obligations apply from 2 December 2027 for stand-alone Annex III systems and from 2 August 2028 for AI embedded in Annex I products, adds two Article 5 prohibitions applying from 2 December 2026, and leaves the 2 August 2026 general application and Article 50 transparency dates in place. Recorded from concordant secondary reporting; confirm against the Official Journal text.
Practical impact: Re-plan high-risk conformity work to the December 2027 and August 2028 dates; transparency duties under Article 50 still applied from 2 August 2026.
Milestone recorded for Artificial Intelligence Basic Plan (Japan): Second AI Basic Plan adopted by Cabinet decision. Recorded from the official source listed.
Milestone recorded for Artificial Intelligence (Malaysia): Ministry of Digital begins engagement and public consultation on the AI Governance Bill. Recorded from the official source listed.
SB 315 became Public Act 104-0538. Large frontier developers must publish a frontier AI framework and transparency reports, report critical safety incidents to the state, protect whistleblowers and, from 2028, commission annual independent third-party audits. Recorded from search extracts of the ILGA bill status page.
Practical impact: Large frontier developers should map their California SB 53 framework to Illinois's audit requirement.
S. 4825 would levy a one-time tax, payable in equity, on AI companies with more than USD 200 million in annual sales, placing the shares in a public fund managed by an independent commission and paying dividends to residents. Introduced and referred; no committee action recorded. Recorded from secondary reporting and the sponsor's press release; confirm on Congress.gov.
Practical impact: No obligations; a proposal to watch in the economic-transition tracker.
Milestone recorded for Draft South Africa National Artificial Intelligence (South Africa): Withdrawal gazetted (Notice 3978 of 2026, Gazette No. 54840). Recorded from the official source listed.
Public Act 26-15 was signed. Its frontier provisions, effective 1 October 2026, protect whistleblowers and require large frontier developers to set up an internal reporting process and report to their boards. Recorded from search extracts of the General Assembly's pages.
Practical impact: Large frontier developers should have an anonymous internal reporting process by 1 January 2027.
Milestone recorded for National Artificial Intelligence Strategy 2.0 (Singapore): Update to the National AI Strategy released with 10 refreshed priorities. Recorded from the official source listed.
The Governor signed SB 26-189, which repeals and re-enacts the Colorado AI Act (SB 24-205) before its delayed 30 June 2026 effective date. The replacement law narrows the scope to automated decision-making technology that materially influences consequential decisions, drops the duty of care, impact assessments, the risk-management programme and Attorney General notification, and adds advance notice, adverse-decision disclosure, meaningful human review, three-year record keeping and developer documentation duties. Recorded from secondary reporting; confirm against the enrolled bill.
Practical impact: Stop building to the SB 24-205 impact-assessment and risk-programme requirements; plan notice, adverse-decision disclosure, human review, record keeping and developer documentation for consequential decisions from 1 January 2027.
Milestone recorded for Rancangan Peraturan Presiden tentang Peta Jalan Kecerdasan Artifisial Nasional 2026-2029 (Indonesia): Government completes inter-ministerial finalisation of the roadmap and ethics draft regulations. Recorded from the official source listed.
Milestone recorded for Draft South Africa National Artificial Intelligence (South Africa): Minister Malatsi announces withdrawal of the draft policy. Recorded from the official source listed.
Milestone recorded for Ghana National Artificial Intelligence Strategy (Ghana): Official launch in Accra; Minister's opening remarks published. Recorded from the official source listed.
Milestone recorded for Plan Nacional de Inteligencia Artificial (Mexico): Plan Nacional de IA presented by ATDT (April 2026). Recorded from the official source listed.
The chapter amendment to the RAISE Act was signed as chapter 96 of 2026. Its sponsor memo says it repeals the underlying 2025 chapter and establishes a new regulatory regime aligned more closely with California SB 53. Recorded from search extracts of the Senate bill page; the effective date was not confirmed.
Practical impact: Frontier developers should read the RAISE Act as amended, not the 2025 text.
Milestone recorded for Law on Artificial Intelligence (Vietnam): Law on Artificial Intelligence enters into force. Recorded from the official source listed.
Milestone recorded for Rancangan Peraturan Presiden tentang Peta Jalan Kecerdasan Artifisial Nasional 2026-2029 (Indonesia): Inter-ministerial plenary discusses draft Perpres on National AI Roadmap 2026-2029. Recorded from the official source listed.
Milestone recorded for National Artificial Intelligence Policy 2026-2030 (Bangladesh): Draft V2.0 published; consultation closed 8 February 2026. Recorded from the official source listed.
Milestone recorded for National Artificial Intelligence Policy 2026-2030 (Bangladesh): Draft V1.1 released for public consultation. Recorded from the official source listed.
IMDA launched a voluntary governance framework for AI agents at the World Economic Forum, covering upfront risk bounding, human accountability, technical controls across the agent lifecycle and end-user responsibility, and said it was developing testing guidelines for agentic applications. Recorded from search extracts of IMDA and MDDI pages.
Practical impact: Organisations deploying AI agents can use it as a reference for governance and controls.
Milestone recorded for Framework Act on the Development of Artificial Intelligence and Establishment of a Foundation for Trust (South Korea): Act enters into force. Recorded from the official source listed.
AI Basic ActAct enters into forceVerified (reviewed by Bhaskar Bhatt; a factual check against the official source, not a legal review)
The Working for Workers Four Act provisions took effect requiring employers with 25 or more employees to state in public job postings whether AI is used to screen, assess or select applicants.
Practical impact: Add an AI-use statement to every public job posting for Ontario roles.
Public Act 103-0804 became effective: discriminatory AI use in employment decisions is a civil-rights violation and employers must notify employees of AI use.
Practical impact: Employers must have notices and bias testing in place for AI tools affecting Illinois workers.
Milestone recorded for National AI Strategy (Egypt): NCAI mandate expanded to AI, Quantum Computing and Emerging Technologies (January 2026). Recorded from the official source listed.
Most provisions of the Transparency in Frontier Artificial Intelligence Act took effect, including the duty to publish a frontier AI framework and report critical safety incidents.
Practical impact: Large frontier developers should now have a published framework and incident-reporting process.
The most recent change on record occurred on 7 November 2026. The page is generated from the records, so it changes the moment a new entry is logged; the "Last updated" time above is the last edit to any record shown.
Where do these updates come from?
Every entry is written from an official source — a journal, a regulator, a legislature — and links to it. Nothing here is a press summary of a press summary. Entries record what changed, the status after the change and what it means in practice.
How is "top story" decided?
By a published rule, not an editor's mood: impact level, whether the instrument is binding, whether it entered into force, whether a named reviewer confirmed the record, and how recent it is, added up into a score that orders the list. Each story shows its recorded impact level: urgent, high or routine.
Can I follow updates for one country?
Yes. Every jurisdiction with recorded changes has its own updates page and RSS feed, and the weekly digest can be limited to the jurisdictions you choose.
Disclaimer: informational only, not legal advice. Verify every claim against the linked official sources and consult a qualified lawyer before acting.
