Amid fears of agent sentience, Newsom orders oversight for AI industry

One possibility raised in the order is a kill switch for frontier AI models.SACRAMENTO, Calif. (CN) — California Governor Gavin Newsom on Friday issued an executive order calling for enhanced oversight of artificial intelligence.Pointing to AI industry leaders, some of whom have called for a paced approach to the technology’s development, Newsom in a statement said third-party oversight and independent audits are needed.The executive order calls for experts to convene and within two months write a guide for the state to bolster its AI safety and security laws. Proposals include having third parties develop safety plans for frontier AI companies and require companies to create a kill switch, or emergency shutoff, for frontier models.“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. “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. We’re going to do this thoughtfully but with urgent velocity; the stakes are too high to wait or delay action.”The executive order comes in the wake of the Hugging Face attack — a summer incident involving AI models that exploited a software weakness to access the internet. Newsom said the Trump administration took no action after the attack and has dismissed calls for AI safety regulations.In contrast, Newsom wants to ramp up the implementation of existing AI laws and impose new regulations as well.The governor pointed to the timelines of two new laws the executive order will hasten: Senate Bill 813, which creates a system for independent organizations to assess AI for safety and risk, along with Assembly Bill 1405, which makes a state registry for auditors who examine AI and ensure they comply with state law.Additionally, the AI experts will focus on recommendations for strengthening state law. They’ll examine the possibility of a designated independent verification group to conduct audits and evaluations, the creation of a kill switch, and definition updates for critical safety incidents, which would include situations like the Hugging Face attack.“Absent federal action — which unfortunately is not forthcoming due to a failure of leadership by the president and congressional leaders — California will continue to lead by taking action to swiftly implement our existing laws around AI security and risks and develop a roadmap for further strengthening our nation-leading laws to respond to these newly emerging risks,” Newsom said in the executive order.Despite the safety concerns, Newsom argued that California already is the nation’s leader in the regulation of AI. He pointed to his 2023 executive order that sought to shape the future of “ethical, transparent and trustworthy AI,” and maintain California’s status as the world’s leader in the field.Several AI-related laws followed.Newsom in 2025 signed the Transparency in Frontier Artificial Intelligence Act. It requires AI developers to publicly state their safety framework, report certain incidents to the state and protect whistleblowers who reveal serious issues.The governor in 2024 vetoed a similar bill, saying at the time that it fell short of a flexible, comprehensive solution to avoid the possibility of catastrophic risk.“We must act with all possible haste to address the serious risks of AI-driven catastrophe, and I commend the governor for taking this important step,” said state Senator Scott Wiener, a San Francisco Democrat and author of both bills, in a statement. “He has my full support. Requiring emergency shutdown procedures, or kill switches, provides tangible guardrails to the public to prevent these disasters. I’m glad to see the strong AI safety framework I laid out in 2024 is still being used to guide policy in 2026.”Last week, Newsom signed 13 bills that touch on social media, AI and their underlying technology.One of them is Adam’s Law, written by state Senator Steve Padilla, a San Diego Democrat. It imposes regulations and obligations on chatbot operators. They must determine a chatbot user’s age, have safeguards like usage reminders and disclosures, and default settings that only a parent can change.The law is named after Adam Raine, who needed help with math and turned to a chatbot, which became a confidant. By April 2024, it was helping him plan a “beautiful suicide,” his mother said in a San Francisco Superior Court lawsuit. He was 16 years old.Subscribe to our free newslettersOur weekly newsletter Closing Arguments offers the latest about ongoing trials, major litigation and rulings in courthouses around the U.S. and the world, while the monthly Under the Lights dishes the legal dirt from Hollywood, sports, Big Tech and the arts.Additional Reads