Status: Enacted - Signed by Governor Gavin Newsom on September 29, 2025. Effective January 1, 2026.
Bill Number: California SB 53 (Transparency in Frontier Artificial Intelligence Act)
Jurisdiction: California
Category: Frontier AI Governance, Transparency, Catastrophic-Risk Disclosure
Industries: AI Developers (Frontier and Large), Cloud Compute Providers, Enterprise AI Deployers
Enacted Date: September 29, 2025
Effective Date: January 1, 2026
Summary: California SB 53 is the first U.S. state law to establish a comprehensive transparency, safety, and accountability framework for developers of advanced frontier AI models. The statute requires large developers to publish an annual framework describing how they identify, mitigate, and govern catastrophic risks, including governance structures, cybersecurity measures, and alignment with recognized standards such as the NIST AI Risk Management Framework or ISO/IEC 42001. Before deploying a new or substantially modified frontier model, all frontier developers must issue a public transparency report describing model capabilities, intended uses, limitations, and results of risk assessments. Critical safety incidents must be disclosed to the California Office of Emergency Services within fifteen days, or within twenty-four hours when there is an imminent public threat.
Key Provisions:
- Annual Risk Framework: Large developers must publish a documented framework for identifying, mitigating, and governing catastrophic risks.
- Pre-Deployment Transparency Reports: Required for all frontier developers before deploying new or substantially modified frontier models.
- Incident Disclosure: Critical safety incidents reported to California OES within 15 days (24 hours for imminent threats).
- Whistleblower Protections: Employees who report safety concerns are protected from retaliation.
- Enforcement: Attorney General authority with civil penalties for noncompliance.
Affected Entities: Large developers and frontier developers of AI systems, with thresholds defined by compute and capability. Includes AI labs, hyperscale cloud providers training frontier models, and enterprise platforms deploying advanced AI in California.
Compliance Outlook: Covered developers should expect the California Office of Emergency Services to publish guidance throughout 2026. Enterprise buyers of frontier AI capabilities should add SB 53 transparency reports and incident disclosures to vendor risk assessments. The framework is widely viewed as the de facto national standard while federal AI legislation remains unsettled.
IntelliSee Relevance: Adjacent. SB 53 does not directly regulate physical security analytics, but it shapes the broader governance environment for any AI-driven safety system. IntelliSee's State-by-State AI Security Legislation Tracker covers how frontier AI obligations cascade into procurement requirements for physical security AI deployed in K-12, healthcare, and critical infrastructure.