AI Frontier Institute

California AI Transparency Act.

View primary source →

California Senate Bill 1000 (Chapter 861) amends the California AI Transparency Act (CATA) by revising several sections of the Business and Professions Code (Sections 22757.1–22757.5) and adding temporary Section 22757.4.1. Key changes include: (1) removing the 1,000,000 monthly user threshold from the definition of 'covered provider,' making it apply to any person that creates, codes, or produces a publicly accessible generative AI system in California (Section 22757.1); (2) replacing 'AI detection tool' with 'disclosure verification tool' and specifying its requirements, including user privacy protections, public accessibility, and API support (Section 22757.2); (3) requiring latent disclosures in AI-generated or AI-altered image, video, or audio content to include whether the GenAI system created or altered the content, and beginning January 1, 2029, whether the system is designed primarily as assistive technology (Section 22757.3); (4) prohibiting covered providers from falsely representing that a GenAI system is designed primarily as assistive technology (Section 22757.3(d)); (5) removing the previous requirement for covered providers to offer users the option to include a manifest disclosure; (6) setting a civil penalty of $5,000 per day per violation for general CATA violations (Section 22757.4), while imposing a heightened penalty of $50,000 per day for falsely representing a system as assistive technology under new Section 22757.4.1 (which is repealed on January 1, 2029); and (7) exempting GenAI systems designed primarily as assistive technology from CATA obligations until January 1, 2029 (Section 22757.5). The bill takes effect immediately as an urgency statute, citing public health, election integrity, and anti-fraud rationales.

← Back to the tracker