California is moving to expand independent safety oversight for artificial intelligence companies operating in Southern California and statewide, potentially requiring an emergency “kill switch” for advanced AI models. Governor Gavin Newsom issued an executive order on Friday to speed up the state’s approach to AI regulation.
The order directs the California Government Operations Agency, in consultation with the Governor’s Office of Emergency Services, to convene national experts. This group is tasked with recommending additional safeguards for advanced AI systems, with proposals expected within two months on how California can strengthen its existing AI safety and security laws.
Among the key proposals to be considered is a requirement for companies developing advanced, or “frontier,” AI models to implement an emergency shutoff mechanism, often called a kill switch. The effectiveness of such a system would then be subject to ongoing verification by an independent organization.
Experts will also explore requiring these frontier AI companies to have independent verification organizations stationed onsite to conduct regular audits and evaluations of their systems. Other potential changes include mandating independent verification of safety frameworks, transparency reports, and risk assessments submitted by AI companies. The state’s definition of critical safety incidents could also be expanded to include incidents involving a loss of control.
Newsom emphasized the urgency of the situation, stating, “the stakes are too high to wait or delay action.” The Governor's office indicated that the action was prompted by recent AI security incidents and concerns raised by researchers and industry leaders regarding increasingly powerful AI systems.
This executive order follows Newsom’s recent signing of Senate Bill 813 and Assembly Bill 1405 last week. SB 813 establishes a framework for certifying independent organizations to verify AI systems’ compliance with state law, while AB 1405 creates a state registry for AI auditors and sets standards for their independence, transparency, and integrity. The new order directs state officials to accelerate the implementation of both these laws.
California already has several AI-related regulations in place. The state requires developers of frontier AI models to publicly disclose safety frameworks and report specified critical safety incidents under legislation Newsom signed in 2025. Additionally, the state has enacted laws concerning AI auditors, companion chatbots used by children, AI-generated content, privacy, and the use of artificial intelligence by state government.
Newsom also called on the federal government to adopt California’s AI regulatory framework, suggesting the state’s standards should serve as a national baseline. He noted that California has already established a national model for AI regulation, highlighting the state's proactive stance as Washington continues to grapple with its own approach to AI oversight.
