California Governor Gavin Newsom issued an executive order Friday convening a panel of national artificial intelligence experts to establish a comprehensive framework reinforcing the state’s frontier safety and security regulations. The directive aims to fast-track independent third-party oversight for advanced artificial intelligence systems, directing the newly assembled expert group to deliver a finalized blueprint within two months.
The executive order marks the latest step in California’s ongoing effort to establish rigorous, proactive safeguards for emerging technologies amid a perceived lack of federal intervention. According to the governor’s office, the accelerated initiative introduces several aggressive proposals designed to mitigate potential risks associated with rapidly evolving frontier models.
Under the newly outlined framework, state officials are considering mandatory requirements for frontier AI developers to embed independent verification organizations directly within their operational labs. These onsite entities would be tasked with conducting continuous audits, safety evaluations, and real-time risk assessments. Additionally, the proposal mandates the strict verification of existing safety frameworks, transparency reports, and risk assessments that frontier AI companies must already file under California state law.
A particularly notable component of the executive order is the advancement of a structural "kill switch" for frontier models, designed to halt runaway or hazardous system behavior in emergency scenarios. Furthermore, the initiative calls for an update to the legal definitions governing critical safety incidents, formally expanding them to encompass catastrophic loss-of-control events.
Governor Newsom sharply criticized the federal government’s approach to technology policy, framing the state-level actions as an urgent necessity to protect the public.
"The federal government’s abject failure to create any form of meaningful AI oversight or accountability should alarm every American, especially when AI CEOs themselves are begging for regulation," Newsom said in a statement released Friday. "We’re not waiting to act; we’re going to speed up our work on substantial and responsible AI oversight before it’s too late."
Friday’s executive order builds upon an extensive catalog of artificial intelligence legislation and regulatory frameworks recently enacted in California. The state has previously established laws covering frontier model safety, independent oversight, privacy protections, deepfake restrictions, fraud prevention, cybersecurity for critical infrastructure, workforce preparedness, and government technology deployment.
Earlier in September, Governor Newsom signed two pivotal bills establishing what California officials describe as the nation’s first statutory framework for independent AI verification organizations and auditors. That legislation also incorporated independent auditing directly into the product safety compliance process for consumer-facing companion chatbots.
The state’s legislative and executive focus extends broadly across various societal impacts of artificial intelligence. In May, Newsom signed a pioneering executive order directing state agencies to prepare workers, local communities, and small businesses for potential economic and labor market disruptions driven by rapid technological advancements.
Children’s safety and consumer protection have also been central pillars of California’s regulatory agenda. In October, Newsom signed landmark legislation designed to protect children and vulnerable users from the psychological and developmental harms associated with AI-driven social media and interactive chatbots, making California the first state in the nation to formally regulate virtual companions.
Reflecting on the need for targeted protections for minors last October, Newsom emphasized the dual nature of emerging digital platforms.
"Emerging technology like chatbots and social media can inspire, educate and connect, but without real guardrails, technology can also exploit, mislead and endanger our kids," Newsom noted at the time.
With the delivery of the new expert guide mandated within sixty days, the state government is positioned to further codify its oversight mechanisms, setting up a continued divergence between California’s regulatory posture and federal policy regarding the governance of artificial intelligence.