AI Regulation Tracker / Reference
Global AI Compliance Deadlines Calendar
The AI regulation deadlines that change a professional's duties, worldwide, ordered by date. We update this as new rules land.
Upcoming deadlines
| Date | Jurisdiction | Who must act | What changes |
|---|---|---|---|
| July 1, 2026 | Australia | Banks, insurers, super funds | APRA CPS 230 takes effect; AI and cloud vendors behind a critical operation must be managed as material service providers with tested fallbacks. Read |
| July 6, 2026 | United Kingdom | Financial firms, senior managers | FCA Mills Review published; the Senior Managers Regime continues to apply, so a named senior manager stays accountable for AI outcomes. Read |
| July 15, 2026 | China | Companion and emotional AI providers | CAC anthropomorphic AI interaction measures take effect; filing, AI-nature disclosure and safeguards required. Read |
| July 31, 2026 | United States | Any business marketing AI | Comment closes on the FTC proposal treating suppressed AI accuracy as a Section 5 deception. Read |
| August 12, 2026 | Colorado, US | Mental-health clinicians | Colorado HB 26-1195 takes effect; AI may not conduct therapeutic communication without real-time clinician oversight. Read |
| August 26, 2026 | Brazil | Physicians | CFM Resolution 2.454/2026 takes effect; physicians stay responsible for AI-assisted decisions and must record AI use in the chart. Read |
| September 10, 2026 | Peru | Health, education, justice, security, economy, finance | First sectoral compliance deadline under Peru's Law 31814 regulation (DS 115-2025-PCM). Read |
| October 2026 | Brazil | Campaigns, parties, platforms | TSE Resolution 23.755 governs the October vote; deepfakes banned, AI content labeled, 72-hour synthetic-content blackout around polling. Read |
| December 2, 2026 | European Union | AI providers and deployers | The AI Act's new Article 5 ban on AI that generates non-consensual intimate imagery or CSAM takes effect (per legal analyses of the adopted text). Read |
| December 2026 | Brazil | AI firms, DPOs | ANPD live AI regulatory sandbox testing phase concludes, informing future binding rules. Read |
| January 1, 2027 | Iowa, US | Health insurers, physicians | Iowa HF 2635 takes effect; AI cannot be the sole basis for a prior-authorization denial and a same-specialty reviewer is required. Read |
Rolling and in-force duties (no single deadline)
| Jurisdiction | Who | Duty |
|---|---|---|
| Canada | Lawyers | Cost sanctions for AI-fabricated citations (Mazaheri; Reddy v Saroya); the verification duty attaches at filing. Read |
| Brazil | Lawyers | The TST labor court fined a lawyer and company for AI-invented citations and referred both to the bar and prosecutors. Read |
| Brazil | Courts, judges | CNJ Resolutions 615 and 674 require human oversight of AI across the courts (12-month compliance). Read |
| Brazil | Controllers using biometrics | ANPD enforcement over facial recognition (23 clubs) and forthcoming binding biometric rules. Read |
| Brazil | Law-firm partners | The Sao Paulo bar makes partners answerable for supervising associates' AI use. Read |
| Brazil | AI developers and firms | The PL 2338 AI Act and its AI-training copyright fight remain pending in the Chamber. Read |
| Brazil | Insurers, insurtechs | SUSEP made its insurance sandbox permanent, the supervised route for AI in claims and underwriting. Read |
| Global | All firms | The first reported AI-run ransomware maps to existing breach, safeguards, and notification duties. Read |
Sponsored Training
Informational analysis for working professionals, not legal advice. Confirm how any rule applies to your situation with qualified counsel.