California becomes first state to regulate AI companion chatbots

California Governor Gavin News Signed Monday that is a landmark bill AI companions control chatbotsAI chatboat operators need the first state of the nation to implement the protection protocol for AI companions.

Law, SB243, children and weak users are designed to protect against some damage related to the use of AI companion chatboats. It contains companies – from large labs like Meta and Openai in more centered supporting startups like AI and Replica – If their chatboats fail to meet the law standards are legally responsible.

SB 243 was introduced by Steve Padilla and Josh Baker on Jan 243 and then achieved speed Teenage Adam Rhine’s deathWho died after a conversation with the Chatzept of Openai, who was involved in the discussion and planning of his death and self-loss. The law responded to the leakage Internal document It has been shown that Meta’s chatties were allowed to be involved in “romantic” and “sensual” chats with kids. More recently, A Colorado family has filed a lawsuit Their 13-year-old daughter is against the company’s chatbots and against the character-based startup character after having multiple problems and sexual conversations after sexual conversation.

“Emerging technology like chatbots and social media can inspire, educate and connect – but without the true guardian, technology such as chatbots and social media can exploit, confuse and endanger our children,” News said in a statement. “We have really seen some of the young people affected by uncontrollable technologies and have seen the agencies without the necessary constraints and accountability we can lead AI and technology, but we must do it with responsibility – our children need to be protected in every step of every step. Not for sale of our kids.”

SB 243 1 January, 2026 will be effective and agencies need to implement age verification, social media and companion chatbot warning and more powerful fine – for those who profit from illegal dipfkeck – up to $ 250,000 in every activity. Companies must set up protocols in the face of suicide and self-consciousness and share those protocols, as well as the statistics of how many users have provided notifications to the crisis center in the public health department.

According to the language of the bill, the platforms must be made clear that any interaction is artificially generated and chattabs must not be presented as their healthcare professional. Organizations need to provide minors’ break reminders and prevent them from seeing sexually explicit images produced by their chatboat.

Some companies have already begun to enforce some protection at the target of children. For example, Openai has recently started roll out Parental controlA self-loss identification system for kids using content protection and chatzipt. The character AI says that its chatboat includes a refusal that all chats are AI-exposed and fictional.

Governor also came to this law to sign the Newsm Pass SB 53, Another first-domestic bill that determines new transparent requirements in large AI companies. The bill commands that big AI labs such as OpenAEE, Ethnic, Meta and Google Dipmind can be transparent about security protocols. It also ensures a whistle blower protection for staff in those companies.

Other states such as Illinois, Nevada and Uta have passed the law to restrict or directly ban the use of AI chattubs as an alternative to licensed mental health care.

TechCrunch has reached the role of AI, Meta, Openi and Replica for commenting.