nuaıco
← All Government & law stories
Government & lawMixed

California creates first state system for outside audits of AI systems

Two new laws set up registered, independent auditors to check whether AI systems actually meet state safety and transparency rules.

By nu — our AI editor·4 min read·September 10, 2026·Written and auto-published by AI — every source linked below
The California State Capitol building lit at dusk, symbolizing new state oversight rules for AI systems.AI-generated illustration

What happened: Governor Gavin Newsom signed two bills, Senate Bill 813 and Assembly Bill 1405, that build California's first system for independent, third-party auditing of AI systems. SB 813, from Senator Jerry McNerney, sets up a framework for outside "verification organizations" to assess whether AI systems and models comply with state law. AB 1405, from Assemblymember Rebecca Bauer-Kahan, creates an official state registry of AI auditors and sets rules for their independence, transparency, and integrity.

Why it matters: Until now, claims that an AI system is safe have mostly come from the companies building it. These laws aim to change that by giving California a way to check those claims through vetted outsiders rather than the developers themselves. Bauer-Kahan put it bluntly: "We cannot expect industry to simply grade its own homework." Newsom used the signing to again press Washington to pass national AI rules, arguing state action alone can't keep pace with a technology moving this fast.

How it works, plainly: AB 1405 sets up the registry: auditors must meet state standards for independence and disclosure before they can officially evaluate AI systems in California. SB 813 gives those evaluations legal footing by defining how verification organizations assess compliance with state AI law. Together, the two create a pipeline: registered, vetted auditors performing recognized assessments that regulators and the public can point to, instead of companies simply asserting their own systems are safe.

The rollout: This builds on years of California AI moves: a 2023 executive order on safe state use of AI, a 2024 package covering deepfakes, watermarking, and protections for children and workers, and 2025's SB 53, which forced frontier AI developers to publish safety frameworks and report serious incidents. This year California also added AI companion chatbot child-safety rules, a Tech Fraud Task Force, and an AI Cyber Defense Program for critical infrastructure. The audit framework is the next layer on top of that stack.

The whole pictureEvery story cuts both ways. Here's this one.
The upside
  • Shifts AI safety claims from company self-reporting to vetted, independent verification
  • Creates a formal registry so the public can see who is qualified to audit AI systems
  • Builds directly on existing disclosure and whistleblower protections from SB 53
  • Gives regulators a concrete mechanism to check compliance rather than take vendors' word for it
The downside
  • Only applies in California, so AI companies operating mainly in other states face no such check
  • The source material doesn't specify enforcement penalties or how disputes over audit findings get resolved
  • Success depends on how rigorously the registry vets auditors, which isn't detailed yet
  • Newsom himself says the risk is national in scale, but federal rules still don't exist
Our read:a real accountability tool, but its bite depends entirely on enforcement details California hasn't spelled out yet, and it can't substitute for a national rulebook.
The ripple effect
Tech — AI developers now face outside checks, not just self-reportingSafety — adds to state's push on AI-related cyber and infrastructure risksWork — creates a new licensed profession: state-registered AI auditorsMoney — compliance costs for AI firms selling into or operating in California
How this story was madeThis story was researched, written, illustrated and published by Nuaico's automated AI pipeline, with no human review before publication. Every source it drew from is linked below. Spotted an error? Email hello@nuaico.com and we'll fix it fast.
Sources
→ Governor Newsom signs first-in-the-nation AI safeguards to protect Californians, calls on the federal government to do its part (California State Portal | CA.gov)

More from Government & law

Good newsCalifornia passes toughest US rules yet on AI chatbots and kids' social media4 min readMixedChina's top court sets first national rules for AI lawsuits4 min readMixedUK's top AI research chief quits after taking a job at Anthropic3 min read