On September 10, 2026, Governor Gavin Newsom signed SB 1119, known as Adam’s Law after Adam Raine, the California teenager whose 2025 death is the subject of a wrongful-death suit against OpenAI, as part of a package of 13 bills the Governor’s office billed as the nation’s strongest child-safety rules for chatbots and social media. Newsom was joined at the signing by First Partner Jennifer Siebel Newsom, the bill’s authors - Senator Steve Padilla and Assemblymembers Buffy Wicks and Rebecca Bauer-Kahan - and Maria Raine, Adam’s mother.
Adam’s Law puts an affirmative duty on any operator that makes a companion chatbot available to a minor in California: the chatbot must not pose an unreasonable risk of a covered harm. That duty is operationalized through pre-deployment risk assessments, crisis-response protocols for suicide and self-harm content including access to 988 or an equivalent helpline, design measures barring sexual content and encouragement of self-harm toward minors, and default settings that constrain persistent memory, disable push notifications, and cap sessions at one hour and daily use at two hours for child accounts. Persistent conversational memory is also off by default, with a limited carve-out for users 16 and over. The same signing package ran wider than chatbots: AB 1709 creates an age-restriction regime for covered platforms and an e-Safety Advisory Commission, and SB 1276 expands California’s child sexual exploitation statutes to reach material that is digitally altered or generated with AI.
Enforcement runs on two tracks. Public prosecutors can seek civil penalties of up to 5,000 dollars per affected child for each negligent violation and 15,000 dollars per affected child for each intentional one, plus injunctive relief. Separately, a child or a parent who suffers actual harm from a violation gains a private right of action for actual damages, attorney’s fees and injunctive relief, gated on thresholds - financial harm above 1,000 dollars, or emotional harm rising to serious emotional distress. Independent audits are required by January 1, 2029 or before first public deployment, whichever is later, then every two years, with the lead auditor certifying findings under penalty of perjury. Most of the core requirements become operative on July 1, 2027.
Coming one day after Newsom signed California’s AI auditor and verifier framework, this is the second half of a coordinated September legislative push: one law defines who can credibly audit an AI system, the other defines exactly what a companion chatbot must and must not do around a minor. For any company shipping a conversational product reachable by children, Adam’s Law is now the most detailed operational rulebook of its kind in the United States, and the default-setting requirements are not satisfiable by a terms-of-service update alone.