California moves to add emergency shutdown authority and cap recursive self-improvement in AI

ShakeelHashim · x · 2026-11-19

California Governor Gavin Newsom announced new steps to strengthen the state's AI safety framework, including mandatory baseline safety standards for all AI developers, independent third-party oversight, emergency shutdown authority for dangerous systems, and new authority to cap recursive self-improvement (RSI) — AI that can build more powerful versions of itself. Newsom criticized the federal government for failing to create meaningful AI oversight, framing California as the nation-leading regulator. It's among the most aggressive state-level AI mandates in the US to date.

Related event: California Governor's Expert Report Urges AI Kill Switch, Mandatory Standards and RSI Compute Caps(8 posts)→

Original post →

More from Safety

Safety channel →