AI Regulation Tracker

What we captured

The proposed Ban Artificial Superintelligence Act would permanently ban development and use of artificial superintelligence, freeze development of advanced AI systems pending safety rules, and establish a cabinet-level federal AI agency.

• • Introduced by Bernie Sanders and Greg Casar on September 23, 2026.
• • Proposes a permanent ban on developing and using artificial superintelligence.
• • Proposes freezing development of advanced AI systems until a new federal agency establishes safety rules.

Source evidence

Confidence 97% / Completeness 95%

What we captured

Executive Order 26-26 establishes AI procurement and safety standards for Oregon’s executive branch and directs development of third-party review criteria for frontier AI models.

• • Executive Order 26-26 took effect immediately.
• • The State Chief Information Officer must submit an implementation proposal within 90 days.
• • The order requires reassessment every three months.

Source evidence

Confidence 98% / Completeness 100%

What we captured

Proposed U.S. legislation would ban artificial superintelligence, pause advanced AI development until guidelines are implemented, and establish a federal AI oversight department.

• • Sanders and Casar are unveiling the bill on Sept. 23, 2026.
• • The bill would ban artificial superintelligence and pause advanced AI development until guidelines are implemented.
• • The bill would create a federal Department of Artificial Intelligence to oversee advanced AI.

Source evidence

Confidence 97% / Completeness 95%

What we captured

The European Commission’s AI Office is analyzing whether AI companions enable illegal child sexual-abuse material or non-intentional intimate images and is treating them as high-concern use cases under the EU AI Act.

• • AI companions are treated as high-concern use cases under the EU AI Act.
• • The analysis concerns compliance with the Act’s prohibitions and general safety obligations.
• • The scrutiny particularly concerns companions built on general-purpose AI models with systemic risk.

Source evidence

Confidence 86% / Completeness 95%

What we captured

New York announced a schedule for covered large frontier-model developers to register and comply with the RAISE Act’s safety, reporting and incident-notification requirements.

• • Hochul announced the implementation schedule on September 21.
• • Developer registration is scheduled to begin in November.
• • Critical safety incidents must be reported to DIGIT within 72 hours.

Source evidence

Confidence 96% / Completeness 95%

What we captured

New York is implementing the RAISE Act, signed by Gov. Kathy Hochul in December 2025. Beginning in November, large AI developers must register with the state and report certain safety incidents within 72 hours.

• • Gov. Kathy Hochul signed the RAISE Act in December 2025.
• • Large AI developers are required to register with New York beginning in November.
• • Covered developers must report certain safety incidents within 72 hours.
• • The article identifies the law as the Responsible AI Safety and Education Act.

Source evidence

Confidence 98% / Completeness 95%

What we captured

New York announced that large frontier AI developers will be directed to register with the state beginning in November 2026 and must meet transparency, safety, incident-reporting, and risk-assessment requirements beginning January 2027.

• • Registration is scheduled to begin in November 2026.
• • Compliance with the law’s strict standards is scheduled to begin in January 2027.
• • Critical safety incidents must be reported within 72 hours.
• • The DIGIT Office will oversee reporting and produce an annual public report.

Source evidence

Confidence 99% / Completeness 100%

What we captured

New York announced planned requirements for AI developers to register with the DIGIT Office, report critical safety incidents within 72 hours, conduct quarterly catastrophic-risk assessments, and provide periodic disclosures under the RAISE Act.

• • Critical safety incidents must be reported to the state within 72 hours, beginning as soon as November 2026.
• • AI developers must register with the DIGIT Office by the new year.
• • Quarterly catastrophic-risk assessments and biennial disclosure statements are required, with disclosures beginning in 2027.

Source evidence

Confidence 94% / Completeness 95%

What we captured

The bill would require NIST to develop national standards for controlling AI agents and calls for government testing of new AI models for national-security purposes before public release.

• • Reps. Josh Gottheimer and Mike Lawler introduced the bipartisan Stop Rogue AI Act.
• • The proposed legislation assigns NIST responsibility for developing national AI-agent control standards.
• • Gottheimer said government should test new models for national-security purposes before they are released publicly.

Source evidence

Confidence 96% / Completeness 95%

What we captured

California Governor Gavin Newsom signed an executive order advancing independent AI auditor oversight, recurring third-party verification, incident reporting, and development of an emergency shutdown framework for powerful AI systems.

• • The executive order moves oversight of AI auditors forward by one year.
• • The order calls for ongoing third-party safety verification and stronger incident reporting.
• • Experts will examine the meaning, technical requirements, and implementation of an AI emergency shutdown mechanism.

Source evidence

Confidence 96% / Completeness 95%