On October 13, the governor of California signed AB 853 to amend the compliance date of the California Artificial Intelligence Transparency Act and to add new obligations for large online platforms. The amendment moves the act’s effective date from January 1, 2026, to August 2, 2026. It also requires covered providers that create, code or produce a generative Artificial Intelligence system to make available a free Artificial Intelligence detection tool that helps users determine whether media content was created or altered by the provider’s generative Artificial Intelligence system, and that outputs any system source data detected in the content.
Beginning January 1, 2027, large online platforms face additional responsibilities. The law defines these platforms as public-facing social media, file-sharing, mass messaging platforms, or stand-alone search engines that distribute content to users who did not create it and that exceeded 2 million unique monthly users during the preceding 12 months. These platforms must detect whether source data, compliant with widely adopted standards, is embedded in or attached to the content they distribute. They must then provide users with a latent disclosure that indicates whether the content was generated or altered by a generative Artificial Intelligence system.
Starting January 1, 2028, capture device manufacturers must offer users the option to include a latent disclosure in content captured by devices first produced for sale in California on or after that date. The required disclosure must communicate the manufacturer’s name, the device name and version, and the time and date when the content was created or altered. Together, the delayed timeline and new requirements establish a phased approach that begins with generation-layer detection tools from providers, extends to platform-level source data checks and user disclosures, and reaches device-level options for embedding standardized disclosures in newly produced hardware.
