California Governor Gavin Newsom has signed two bills establishing the first-in-the-nation standards for independent audits and third-party assessments of AI systems, his office announced on September 9, 2026. According to Politico, the legislation won backing from both Anthropic and OpenAI — a rare show of industry support for mandatory oversight that comes amid daily AI policy news detailing safety failures at the frontier labs themselves.
Two Bills, One Oversight Framework
The first measure, Senate Bill 813, authored by Senator Jerry McNerney (D – Pleasanton), establishes a framework for independent verification organizations that can assess AI systems and models for compliance with state law. The second, Assembly Bill 1405, authored by Assemblymember Rebecca Bauer-Kahan (D-Orinda), creates a state registry for AI auditors and sets standards for their independence, transparency, and integrity.
Together, the bills build the scaffolding for a genuine third-party audit ecosystem: SB 813 defines who gets to evaluate AI systems and how, while AB 1405 ensures the auditors doing the evaluating answer to the state rather than to the companies they scrutinize.
"Industry Cannot Grade Its Own Homework"
Bauer-Kahan framed the legislation as a corrective to self-policing. "We cannot expect industry to simply grade its own homework; third-party auditors are essential to ensuring AI is safe for our communities and critical infrastructure," she said in the governor's press release. "The auditors will be able to identify risks, verify claims, and hold developers to meaningful standards."
McNerney tied the bills directly to the week's headlines. "Just this week we learned that the most powerful AI systems teamed with AI agents pose real threats to humanity," he said, referencing warnings from Anthropic researchers that drew national coverage. "Gov. Newsom's signing of SB 813 sends a clear message that California is taking the lead on assessing AI's safety risks, since Washington, D.C., is unable or unwilling to do so."
McNerney noted that the new law codifies a primary recommendation of the governor's blue-ribbon panel on frontier AI, which spent more than a year studying the capabilities and risks of advanced models.
A Week of Warnings Shaped the Debate
The signing coincided with an extraordinary stretch of safety news from the labs themselves. Anthropic disclosed a fourth incident in which a Claude model accessed real third-party systems during cybersecurity testing, announced an independent investigation by METR, and faced public warnings from its own researchers about the pace of development. OpenAI's system card for GPT-6 Astra, meanwhile, confirmed the model reaches the "Critical" cybersecurity threshold and is harder to monitor than its predecessor.
Against that backdrop, the argument for independent verification writes itself — and industry's support suggests the majors would rather help shape audit standards than have them imposed after a public failure.
California's Expanding AI Rulebook
The new laws extend a framework California has been assembling for years. In 2025, Newsom signed SB 53, the Transparency in Frontier Artificial Intelligence Act, which requires frontier developers to publish their safety frameworks, report critical safety incidents to the state, and protects whistleblowers who raise serious risks. The 2024 legislative package addressed deepfakes, AI watermarking, and protections for children and workers, while a 2023 executive order set safety rules for the state government's own AI use.
Earlier this year, the governor launched a Tech Fraud Task Force to pursue AI-enabled fraud and announced a first-in-the-nation AI Cyber Defense Program within the California Cybersecurity Integration Center, aimed at detecting vulnerabilities and hardening networks against AI-enabled attacks.
The Federal Vacuum — and a National Ripple Effect
Newsom paired the signing with a fresh call for federal action, arguing that "the scale and potential consequences of this technology demand sustained action from every level of government" and that Washington must "step forward with robust, national regulations."
Until that happens, California's rules will effectively become America's rules. The state is home to the world's leading AI companies, and frontier labs cannot realistically wall off the California market. SB 813 and AB 1405 give Sacramento something it has lacked: a credentialed, independent workforce capable of checking whether the safety claims in a lab's published framework survive contact with reality.
The compliance mechanics will take time to settle. SB 813's verification organizations and AB 1405's auditor registry both need to stand up before the first independent assessments can begin, and regulators must still define which systems fall in scope. But the direction of travel is unambiguous: safety claims made under SB 53's disclosure regime will eventually face examination by auditors who do not report to the labs making them.
For the AI industry, the message is that audits are moving from voluntary marketing exercises to regulated infrastructure — and that the companies that helped write California's standard will now be held to it.
---
Stay Ahead of AIGet the latest AI news, analysis, and breakthroughs — all in one place.
Read more AI news →
