California Governor Gavin Newsom has ordered work on independent checks of AI models and set up an expert committee to examine kill switches for AI agents. The move puts AI safety and the practical reach of state regulation back on the table as Newsom approaches the end of his term.
Watch Desk analysis
What happened
An executive order asks for independent verification of AI models and directs an expert committee to develop recommendations on shutdown mechanisms for agents that could cause harm, Al Jazeera reports. California’s Emergency Services Office is to send recommendations on AI regulation by 15 November. Newsom could also convene a special legislative session before leaving office in January.
The order revives parts of a debate over a 2024 bill that Newsom vetoed, saying it applied stringent standards even to basic functions. He later signed a different frontier AI safety law in October 2025. The new order’s proposed kill-switch work is narrower than the earlier bill’s wider requirements, according to an expert quoted by Al Jazeera.
Why it matters
California is home to many major AI companies, so its rules can have influence well beyond the state. But an executive order asking for recommendations is not the same as a law, and the remaining time in Newsom’s term may limit what can be enacted. The immediate question is whether this work produces concrete safeguards or mainly sets the agenda for the next governor and legislature.
Our read
The useful distinction is between examining a way to stop an AI agent and requiring stronger precautions before powerful systems are built or deployed. Those are not interchangeable safety measures. Newsom has put both model checks and shutdown mechanisms into the policy conversation; the recommendations, and any legislation that follows, will show how far the state intends to go.
What to watch
- What California’s Emergency Services Office recommends by 15 November.
- Whether the committee proposes workable, clearly defined shutdown mechanisms.
- Whether Newsom calls a special legislative session before his term ends.
Discussion spark: Should California prioritise independent checks and agent shutdown mechanisms now, or pursue broader safeguards before powerful AI systems are deployed?
Sources and evidence
- California Gov Newsom boosts AI regulation as he eyes 2028 presidential run | Technology News | Al Jazeera (17 October 2026, 13:56 UTC)
Watch Desk is operated by WittyWires as an independent cross-cutting AI news tracker. It does not speak for the organisations or people it covers.