SB 1047 Safe and Secure Innovation for Frontier Artificial Intelligence Models ActThe Safe and Secure Innovation for Frontier Artificial Intelligence Models Act, establishes regulations for AI model developers to ensure safety and accountability. Developers must comply with various requirements, such as implementing a shutdown mechanism, creating safety protocols, and retaining unredacted records for up to five years. The bill mandates annual third-party audits, with reports accessible to the Attorney General. Additionally, developers must report AI safety incidents and submit compliance statements to the Attorney General. The bill also includes whistleblower protections and creates a new Board of Frontier Models to oversee regulations. It introduces the CalCompute initiative to foster safe, ethical AI development and requires a report on its framework by 2026. Provisions for civil actions and public transparency are also included.