Gov. Gavin Newsom on Friday signed an executive order directing California’s Government Operations Agency to accelerate implementation of SB 813 and AB 1405, the two AI auditor statutes he signed only last week. The compliance clock on California’s independent AI auditor mandate just compressed from years to weeks.

The order convenes an expert panel with two months to file recommendations, due Nov. 16. Per the governor’s office, the proposals on the table include onsite independent auditors embedded inside frontier AI labs, third-party verification of safety frameworks, and a mandated “kill switch” for frontier models. The order also broadens what counts as a “critical safety incident” to sweep in loss-of-control events like the Hugging Face hack.

That’s the frontier-lab surface. Underneath sits the auditor registry SB 813 and AB 1405 actually build, and it doesn’t stop at OpenAI’s door.

Any business deploying AI systems to reach California customers, chatbots, personalization engines, generative ad tooling, is downstream of the same infrastructure. Small teams that layered AI on top of their sales and marketing stack this year are now watching the audit and verification standards get written on a 60-day sprint. Combined with the Adam’s Law chatbot risk-assessment regime, the compliance surface for a five-person shop selling into California has quietly become larger than the compliance surface a mid-sized SaaS company faced in 2023.

CalMatters reports Newsom also floated a special legislative session on AI this week in a Politico interview. That’s the second signal in a week that Sacramento intends to treat AI oversight the way it treated data privacy after 2018: fast, unilateral, and structurally binding on anyone selling into the state. The frontier-lab framing is the headline. The auditor plumbing is the story.

Sources

Sources