California Enacts First US Laws Mandating Independent AI Audits and Auditor Registry

California has established the first regulatory framework in the United States for independent AI audits, marking a shift from industry self-regulation to structured oversight1

1

2

2

. Governor Gavin Newsom signed SB 813 and AB 1405 into law on Wednesday, creating what officials describe as nation-leading standards for third-party audits of AI systems3

3

.

California AI Laws Establish Independent Verification Framework

Senate Bill 813, authored by state Senator Jerry McNerney, creates a framework for independent verification organizations to assess AI systems and models for compliance with state law2

2

. Under the legislation, the California Government Operations Agency must establish criteria for independent verification organizations by Jan. 1, 20284

4

. Applicants will need to disclose their qualifications, methodologies and proposed testing tools, while demonstrating sufficient technical expertise and adequate systems for managing conflicts of interest4

4

.

Assembly Bill 1405, authored by Assemblymember Rebecca Bauer-Kahan, establishes a state registry for AI auditors and sets standards for their independence, transparency and integrity2

2

. California must establish an online AI Auditor Registry by Jan. 1, 2029, after which unregistered persons or organizations generally cannot offer, sell or conduct an AI audit required to assess compliance with state law4

4

.

AI Risk Management Through Third-Party Audits of AI Systems

The California AI laws are designed to prevent AI companies from being solely responsible for evaluating their own technology3

3

. "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," Bauer-Kahan said3

3

.

Registered auditors must provide clients with reports describing the audit's scope, objectives, findings and supporting documentation4

4

. Reports must also identify deficiencies, suggest appropriate remedies and disclose limitations, including material gaps in evidence, information, systems or access4

4

. An auditor may accept reasonable payment from the company being assessed, but compensation cannot depend on the findings4

4

.

AI Safety Standards Gain Industry Support from OpenAI and Anthropic

The Gavin Newsom AI bills received endorsements from major AI companies. Anthropic endorsed the legislation, while OpenAI announced its support in the hours before Newsom's signing1

1

. "We've reached a new chapter in AI capabilities, and that demands a new chapter for AI policy," says the introduction of OpenAI's statement, credited to Chris Lehane, OpenAI's Chief of Public Affairs1

1

.

OpenAI stated it "will advocate for compatible international approaches to measuring capabilities, managing risk, preserving human control, and determining when and how development should slow or stop, even if that means slowing the advancement of model capabilities"1

1

. Anthropic pointed to its history of creating safeguards and said "the world would benefit from the industry adopting a lawful, verifiable way to work together to pace how we release powerful models"1

1

.

AI Regulation Expands with Companion Chatbot Law

Newsom signed a companion law, SB 1119, known as Adam's Law, connecting independent audits to specific product categories4

4

. Beginning July 1, 2027, operators must assess child-safety risks before offering a new or substantially modified companion chatbot in California4

4

. Operators must arrange an initial independent child safety audit by Jan. 1, 2029, or before first making a chatbot available, whichever comes later4

4

.

Accountability and Transparency Requirements for AI Auditor Registry

The AI auditor registry includes strict conflict-of-interest provisions. Auditors cannot evaluate AI systems, processes or controls they materially designed or operated4

4

. An employee generally cannot audit an area for which that person held material responsibility at the client during the previous 12 months4

4

. Violations can result in removal from the registry and referral to the attorney general or another enforcement authority4

4

.

The legislation directs the agency to align its criteria with national and international audit and assurance standards where practical4

4

. This provision could help move risk assessments of AI models closer to the controls used in financial reporting, where independence, documented methods and sufficient evidence underpin credibility4

4

.

Gavin Newsom emphasized that while California has moved aggressively on AI regulation, federal action remains necessary. "California has taken nation-leading action to advance AI safety, transparency, accountability, and responsible innovation, but the scale and potential consequences of this technology demand sustained action from every level of government. The federal government must step forward with robust, national regulations that match the urgency of this moment," the governor said3

3

. California has also enacted other measures including SB 53, which requires developers of advanced AI models to publicly disclose safety frameworks, report critical safety incidents and provide whistleblower protections3

3

.