California has become the first US state to pass a law to regulate AI chatbot companions. Governor Gavin Newsom signed SB 243, which requires companies that create such services to implement mechanisms to protect children and vulnerable users. This is reports the official website of the California government.
The new rules include mandatory age verification, risk warnings, and a ban on bots impersonating medical professionals or distributing obscene content. Platforms are required to clearly indicate that the user is communicating with AI, as well as to develop protocols for responding to suicide cases.
If the new law is violated, fines of up to $250,000 are provided for profits from illegal deepfakes. SB 243 will take effect on January 1, 2026, and companies are required to report compliance to the California Department of Health.
The bill specifically aims to prevent companion chatbots — defined as artificial intelligence systems that provide adaptive, human-like responses and are capable of meeting the user’s social needs — from engaging in conversations about suicidal thoughts, self-harm, or sexual content.
The document also requires platforms to provide users with periodic notifications — every three hours for minors — reminding them that they are interacting with an artificial intelligence, not a real person, and that they should take a break.
In addition, the law introduces annual reporting and transparency requirements for companies that create companion chatbots. They will apply to developers such as OpenAI, Character.AI and Replika, and will take effect on July 1, 2027.
