Tracking 100 regulations across 24 states and jurisdictions.

AI Regulations Tracker

Every AI and automated decision-making regulation XIRA tracks, organized by state. Every entry verified against enrolled statute text.

Upcoming effective dates and regulatory changes

Sorted by what arrives soonest, with timeline context and weekly update notes.

Signal: Next deadline is Colorado ADMT / AI Act (SB 26-189) in 81 days.

4 regulations take effect within 90 days

  • January 1, 2027: Colorado ADMT / AI Act (SB 26-189)
  • January 1, 2027: CCPA/CPRA Automated Decision-Making Technology Regulations
  • January 1, 2027: Washington AI Chatbot Safety for Minors (HB 2225)
  • January 1, 2027: New York Responsible AI Safety and Education Act (RAISE Act, S6953B/A6453B)
Run a scan to see if this applies to you

Coming up

5
TODAY

Already in effect

95

These deadlines apply differently depending on your AI tools, your states, and your role. Run a free scan to see which ones matter for your company.

Start your free scan

Regulatory updates

Upcoming in the next 90 days

4 regulations take effect in the next 90 days.

Filters

January 2027

  • January 1, 2027High impactEffective dateCO

    Colorado ADMT / AI Act (SB 26-189) takes effect

    Colorado SB 26-189 repeals and reenacts SB 24-205 into an automated decision-making technology (ADMT) framework for consequential decisions. Starting January 1, 2027, covered developers may need to provide deployers with technical documentation and material-update notices. Covered deployers may need point-of-interaction notices, post-adverse-outcome disclosures, data-access and correction processes, human-review and reconsideration workflows, and three-year compliance records. SB 24-205 risk-management, impact-assessment, and reasonable-care artifacts remain useful governance evidence, but they are historical or reusable controls rather than standalone current-law duties under the new Colorado framework. Effective in about 81 days, so teams should map obligations and prepare evidence now.

  • January 1, 2027High impactEffective dateCA

    CCPA/CPRA Automated Decision-Making Technology Regulations takes effect

    California's ADMT regulations require businesses using automated decisionmaking technology for significant decisions (employment, finance, housing, education, healthcare) to provide pre-use notices, offer opt-out rights, respond to access requests, and conduct risk assessments with annual CPPA filing under penalty of perjury. Effective in about 81 days, so teams should map obligations and prepare evidence now.

  • January 1, 2027Medium impactEffective dateWA

    Washington AI Chatbot Safety for Minors (HB 2225) takes effect

    First-in-nation law requiring AI chatbot operators to disclose AI nature at regular intervals (every 3 hours for adults, every hour for minors) and implement safety measures to protect minors from manipulation, explicit content, and emotional exploitation. Includes self-harm and crisis protocols. Targets conversational AI engagement patterns specifically. Effective in about 81 days, so teams should map obligations and prepare evidence now.

  • January 1, 2027Medium impactEffective dateNY

    New York Responsible AI Safety and Education Act (RAISE Act, S6953B/A6453B) takes effect

    New York's RAISE Act regulates frontier AI model developers. Requires publication of a frontier AI framework, quarterly confidential risk assessment filings with the DFS Office, transparency reports, 72-hour incident reporting, and registration with the DFS Office of Frontier AI Model Developer Transparency and Reporting. Applies to developers with $500M+ revenue operating models above 10^26 FLOP. Chapter amendment (S8828) signed March 2026 is the final version. Effective in about 81 days, so teams should map obligations and prepare evidence now.

May 2026

  • May 19, 2026High impactEffective dateFEDERAL

    TAKE IT DOWN Act (S. 146) takes effect

    Dual-effective-date framework now reaches the platform compliance milestone: criminal provisions already took effect at signing, and platform duties are now in force. The law covers both authentic NCII and AI-generated deepfakes, includes FTC jurisdiction over nonprofits, and requires operational notice-and-removal workflows with statutory timelines.

  • May 14, 2026High impactAmendedCO

    Colorado SB26-189 signed, replacing SB24-205 AI Act framework

    Colorado SB26-189 repeals and reenacts the SB24-205 AI provisions as a covered ADMT framework. XIRA now treats SB26-189 as the current Colorado source version while preserving SB24-205 artifacts as historical or reusable governance evidence.

April 2026

  • April 8, 2026High impactDatabase updateALL

    Database reviewed and expanded

    XIRA regulation database was reviewed and expanded with recent state and federal items. Coverage now includes additional California, New York, Connecticut, Utah, and federal entries reflected in the live tracker.

March 2026

  • March 11, 2026Medium impactEnforcementFEDERAL

    Federal commentary on state AI laws published

    Commerce commentary discussed implementation burden in certain state AI statutes. No federal action from this commentary overrides enacted state obligations.

January 2026

  • January 1, 2026High impactNew lawCA

    California 2026 AI statute wave takes effect

    Multiple California laws in the XIRA database move into force, including frontier model transparency and healthcare disclosure requirements. Teams operating in California should verify notice, disclosure, and governance artifacts before launch windows.

December 2025

  • December 11, 2025Medium impactEnforcementFEDERAL

    DOJ AI litigation posture strengthened

    Federal enforcement signaling from DOJ indicates stronger cross-agency focus on AI-related misconduct and harms. This raises the importance of retained evidence and governance controls across high-risk use cases.

100Regulations tracked
24Jurisdictions
88In effect
OnePath from scan to export

Numbers reflect the live catalog and site configuration.

See the match before you commit

Lead with the scan. Add a work email only if you want updates. Nothing here blocks the free exposure path.