California signs SB 243, the Companion Chatbot Law, regulating AI chatbots

California Senate Bill 243 is a state law regulating operators of AI companion chatbot platforms, requiring disclosures that users are talking to an AI and safety protocols for suicide and self-harm content.

On October 13, 2025, Governor Gavin Newsom signed into law California Senate Bill 243, which was chaptered as Chapter 677 of the Statutes of 2025.

What it is

The law adds Chapter 22.6 to the California Business and Professions Code to cover artificial intelligence used for companion-style interaction.

It applies to any operator making a companion chatbot platform available to users in California, with some duties taking effect immediately and others phased in by July 1, 2027.

The statute excludes bots used only for customer service, a business's operations, or technical assistance, among other carve-outs.

What it requires

If a reasonable person interacting with a companion chatbot would be misled into believing they are talking to a human, the operator must issue a clear and conspicuous notification that the chatbot is artificially generated and not human.

For users the operator knows are minors, the chatbot must provide a notification at least every three hours during continuing interactions reminding the user to take a break and that the chatbot is not human.

Operators must also institute reasonable measures to prevent the chatbot from producing sexually explicit visual material or directly telling a minor to engage in sexually explicit conduct.

An operator cannot let a companion chatbot engage with users unless it maintains a protocol for preventing the production of suicidal ideation, suicide, or self-harm content to the user.

Beginning July 1, 2027, operators must annually report to the Office of Suicide Prevention certain things, including crisis-referral counts and prevention protocols, and the office will post the data on its website.

A person injured by a violation may bring a civil action for injunctive relief and damages equal to actual damages or one thousand dollars per violation, whichever is greater.

What security leaders should do

Teams operating or embedding conversational AI features that could meet the statutory definition should confirm whether they are an operator before the law's January 1, 2026 effective date.

Security and product teams should map minor-detection logic, three-hour break reminders, and self-harm escalation paths against the statute's notification and protocol requirements, and publish the required protocol details on a public-facing page. Logging and retention processes should be built now so the first annual filing to the Office of Suicide Prevention can be produced accurately after the July 1, 2027 reporting start date.

Given the private right of action and per-violation damages, legal and security teams should treat disclosure and crisis-protocol gaps as compliance risk rather than a design choice. Connecticut's Public Act 26-15, effective January 1, 2027, sets a similar companion-chatbot safety bar but enforces it solely through the state Attorney General, with no private right of action of its own.