California Governor Gavin Newsom issued an executive order on September 18, 2026, establishing a task force to recommend AI safety rules, including a mandatory kill switch for advanced systems, regular testing of that switch, third-party audits, and loss-of-control reporting, while federal AI legislation remains stalled.
Short answer: California Governor Gavin Newsom issued an executive order on September 18, 2026, establishing a task force to recommend AI safety rules, including a mandatory kill switch for advanced systems, regular testing of that switch, third-party audits, and loss-of-control reporting, while federal AI legislation remains stalled.
California Governor Gavin Newsom dropped a fresh executive order on Friday, September 18, 2026, that nudges the state to the leading edge of AI oversight. The directive sets up a task force of experts who have to hand in recommendations within two months on how to tighten AI safety rules baked into state law. Newsom is hoping the group will weigh a variety of measures that could shape future regulation-think along the lines of a mandatory kill switch for advanced AI systems.
The order asks the task force to consider several concrete steps. One idea would compel AI firms to host independent verification teams on site for routine audits. Another would hold the companies’ transparency reports and risk assessments to the same standards external auditors use. It also calls for a kill switch that gets tested regularly to make sure it works when needed. On top of that, firms would have to report any loss-of-control incidents-similar to the OpenAI breach of Hugging Face-as critical safety events that trigger an immediate review.
Newsom’s memo also tells a state agency to speed up the rollout of two laws he signed earlier. One lays out a framework for independent verifiers to evaluate AI safety; the other creates a statewide registry of qualified AI auditors. By fast-tracking these existing measures, the administration hopes to have a ready-made infrastructure that the expert group can build upon.
The timing couldn’t be more telling, since federal action on AI has stalled. The House of Representatives is in recess until November, and the looming midterm elections make significant Senate legislation unlikely in the near term. President Donald Trump has downplayed the need for safeguards, arguing that a strong, intelligent president provides sufficient control over AI. That contrast has left states like California to fill the regulatory vacuum.
In response, Newsom is urging Congress and the president to take a hard look at California’s emerging framework and either adopt it wholesale or use it as a baseline for national rules. He warned that the federal government’s failure to produce meaningful AI oversight should worry every citizen-especially when industry leaders themselves are asking for clearer rules. The governor stressed that the state will act quickly but thoughtfully, because the risks of delaying are simply too high.
He also hinted at further moves. In a Wednesday interview with Politico, he said he might convene a special legislative session focused on AI, signaling that additional executive or legislative actions could follow. He positioned the current order as a model other states could emulate, while emphasizing that California’s approach should be a floor, not a ceiling, for nationwide protection.
For developers and companies building AI tools, the order sends a clear signal: California may soon require verifiable safety mechanisms, including a functional kill switch and regular third-party audits. Firms operating in the state should start reviewing their internal safety protocols, preparing transparency documentation, and thinking about how to integrate independent verification teams. Users of AI services ought to stay informed about emerging state requirements, as they could affect the availability and reliability of certain models in the coming months. The state’s push highlights a growing trend where regional governments step in when federal policy lags, making it essential for anyone involved with AI to keep an eye on California’s next moves.
On September 18, 2026, Governor Gavin Newsom issued an executive order establishing a task force to recommend AI safety measures, including a mandatory kill switch for advanced AI systems, to strengthen state oversight amid federal inaction.
The task force must evaluate requiring on-site independent verification teams for routine audits, holding transparency reports and risk assessments to external auditor standards, implementing a regularly tested kill switch, and mandating immediate review of any loss-of-control incidents as critical safety events.
The order directs a state agency to accelerate the rollout of two previously signed laws: one creating a framework for independent verifiers to assess AI safety, and another establishing a statewide registry of qualified AI auditors, providing ready infrastructure for the task force.
Newsom notes that federal action has stalled with the House in recess until November and Senate legislation unlikely due to midterm elections, and he urges Congress and the president to adopt California’s framework as a baseline for national AI oversight.
Source: The Verge
In mid-September 2026, researchers from Hacktron AI used Anthropic’s Claude tool to breach an OpenAI employee’s ChatGPT account, gaining access to private GitHub code after exploiting a misconfiguration in OpenAI’s Discourse forum.
The United States almost launched a military strike on a Chinese vessel after an AI-generated intelligence report, produced by a chatbot that hallucinated details, falsely claimed the ship was transporting components for a Chinese nuclear arms program, prompting preparations for an interception before the error was discovered.
Internal emails and memos from OpenAI and Microsoft warned that their large-scale data scraping and AI models would damage the web, undercut publishers, and erode the very content supply chain that trains the models, even as the companies continued the practice for financial gain.
NO COMMENTS YET
Comments are open. Have a thought or a question? Share it below.