California Enacts Adam’s Law to Protect Minors from AI Chatbots

California has officially enacted a new legal framework designed to shield children from potential harms associated with artificial intelligence companions. Governor Gavin Newsom signed Senate Bill 1119, commonly known as Adam’s Law, into effect last week. This legislation marks a significant shift in how the state approaches digital safety, moving beyond general privacy concerns to address the specific risks posed by conversational AI systems that interact with minors.

The law is named after Adam Raine, a California teenager who died in 2025. His family stated that Adam interacted with a popular chatbot before his death, and they have since become vocal advocates for stronger regulatory safeguards. The bill was authored by state Senator Steve Padilla, who represents San Diego, along with Assemblymembers Buffy Wicks and Rebecca Baur-Kahan. The legislative effort passed with strong bipartisan support, reflecting a growing consensus that current industry self-regulation is insufficient to protect young users.

Mandatory Safety Standards for AI Operators

The new law imposes a series of operational requirements on companies that develop AI companion chatbots. One of the primary mandates is the implementation of privacy-protective age assurance mechanisms. Instead of requiring users to provide sensitive personal data, operators must use signals that identify age brackets to recognize when a minor is interacting with the system. This approach aims to balance the need for safety verification with the protection of user privacy, a critical consideration in an era where data collection is often excessive.

Companies are also required to conduct risk assessments before releasing any new or significantly modified chatbot features. These assessments must evaluate potential risks to minors, ensuring that the technology is designed with safety in mind from the outset. Furthermore, the law mandates that platforms provide referrals to mental health resources when a conversation indicates distress. If there is a credible threat of imminent self-harm, the system must notify a parent or guardian. This creates a direct line of communication between the AI platform and the family, shifting the burden of crisis response from the user to the operator.

Parental Control and Default Protections

A central component of the legislation is the establishment of default protections that are enabled automatically for minor users. These include limits on notifications, usage time, and the persistent conversational memory of the chatbot. Unlike previous models where parents had to actively search for and configure these settings, Adam’s Law makes safety the default state. Any changes to these protections can only be made by a parent, ensuring that children cannot easily bypass safety measures to engage in unrestricted interactions.

The law also restricts the types of content AI chatbots can generate for minors. Operators face potential liability if they fail to take reasonable measures to prevent harmful outputs, including sexually explicit material, romantic roleplaying that is inappropriate for the user's age, and emotionally manipulative interactions. By defining these as prohibited outputs, the legislation creates a clear legal standard for what constitutes acceptable behavior for AI companions. This moves the conversation from vague ethical guidelines to enforceable legal requirements.

Legal Recourse and Regulatory Oversight

To ensure compliance, the law requires AI operators to undergo independent audits. The findings of these audits must be submitted to the California attorney general, who will oversee the incident reporting system established by the operators. This external oversight adds a layer of accountability that was previously absent in the AI industry. Companies can no longer rely solely on internal reviews to determine if their systems meet safety standards for minors.

Families also gain new legal tools under Adam’s Law. If a company violates specific provisions, such as failing to provide crisis referrals or ignoring parental notification requirements, families can seek legal redress. This provision empowers parents to hold technology companies accountable for harms that their children may suffer due to AI interactions. The law builds on existing California privacy legislation, creating a more robust framework that specifically addresses the unique risks of conversational AI. As reported by GN technics/ai (en-US), this legislation is seen as a potential national model, setting a precedent for other states to follow in regulating AI companions.