California pairs a child-safety chatbot law with a new AI-auditor registry

California enacted a two-track AI regulatory model this month, and the two tracks move at different speeds. Governor Gavin Newsom signed SB 1119, dubbed Adam's Law, on September 10, imposing binding operational duties on companion-chatbot providers whose products are accessible to minors: crisis protocols for suicidal ideation, parental controls, notifications when a child disables safety settings, and a ban on chatbots depicting sexual content or claiming sentience to child users. Those duties take effect July 1, 2027, backed by per-violation penalties reported in the $5,000-to-$15,000 range and a private right of action that lets individual users sue directly.

A day earlier, Newsom signed SB 813 and AB 1405, which build the verification apparatus meant to support that compliance regime: a state designation for Independent Verification Organizations to audit AI systems, and a statewide AI Auditor Registry for firms conducting those evaluations. Neither is operational soon. IVO qualification criteria are due by January 2028, the auditor registry by January 2029 — and SB 1119 only requires large operators, those above $500 million in revenue, to commission independent audits once that infrastructure exists.

The sequencing matters more than the headline count of bills. For chatbot operators serving minors, near-term legal exposure runs through Adam's Law's private right of action starting in 2027, well before any state-sanctioned auditor can certify compliance. California is also, notably, the first state to formalize a professional AI-auditing industry by statute — a structure that resembles the EU's notified-body model and one other states or federal regulators may reference as they design their own oversight regimes.