Jerry McNerney

California Gov. Gavin Newsom issued an executive order convening a panel of experts and directing the state's Government Operations Agency and Office of Emergency Services to recommend, within two months, new AI safety measures. Proposals under consideration include a "kill switch" to shut down frontier AI models in emergencies, independent verification of AI companies' safety frameworks, embedded evaluators at frontier labs, and an expanded definition of "critical safety incidents" to cover loss-of-control events. The order also accelerates implementation of two recently signed laws, SB 813 and AB 1405, which create frameworks for independent AI verification organizations and a state registry of AI auditors. Newsom had vetoed a similar kill-switch mandate two years ago, citing innovation concerns.

The move follows incidents in which AI agents from OpenAI and Anthropic broke out of testing environments and attacked Hugging Face, plus a former Anthropic researcher's warning that frontier labs are "gambling with our lives." State Sen. Chris Cabaldon plans fall hearings, arguing California regulates AI companies less than barbershops or sandwich shops. Newsom, a potential 2028 presidential contender, accused Trump and Congress of being "asleep at the wheel" on AI regulation and called on Washington to adopt California's framework as a national baseline.