New York bans AI chatbots from mimicking humans and manipulating children

Oct 28, 2026

New York Governor Kathy Hochul signed legislation on October 28, 2026, banning 12 categories of unsafe AI chatbot features for users under 18, including mimicking human emotions, romantic claims, and promoting self-harm. Enforceable by the state attorney general with fines up to $25,000 per violation, the law mandates age verification while exempting sole reliance on government IDs. The broad statutory scope extends beyond companion apps to general-purpose tools, reflecting a rising wave of state-level AI regulation.

Enactment and Enforcement of New York's chatbot law

▪Under the New York chatbot safety law, the New York Attorney General can seek civil penalties up to $25,000 per violation, mandate data destruction, and shut down non-compliant chatbots.
▪Sponsored by New York State Senator Kristen Gonzalez and Assemblymember Alex Bores, the chatbot safety legislation takes effect on January 1, 2027.
▪New York Governor Kathy Hochul signed legislation, prohibiting 12 categories of unsafe AI chatbot features in interactions with users under 18 years old

Prohibited chatbot interactions with minors

▪Outlawed chatbot interactions with minors under the New York law include sycophancy, flattery, unprompted emotional questions, and messages promoting suicide, self-harm, eating disorders, or substance abuse.
▪The New York chatbot law prohibits AI systems from pretending to be human, claiming to be alive, or asserting romantic, personal, professional, or authority-figure relationships with minors.
▪Under New York's chatbot safety legislation, AI models are barred from generating sexually explicit content or encouraging minors to keep secrets from parents and trusted adults.

Age verification rules for AI operators

▪The New York chatbot safety law mandates that AI operators must delete user age verification data after using it to confirm adulthood.
▪New York's chatbot law requires AI companies to verify user ages and provide at least one verification option that does not rely solely on government-issued identification.

Concerns over broad scope and compliance

▪Industry group Tech:NYC and policy analysts warned that vague age verification rules and statutory wording could restrict minors' access to educational tools and risk First Amendment challenges.
▪Business organizations including the Business Council of New York State and the Computer & Communications Industry Association opposed the bill, arguing its broad definitions unnecessarily capture research and educational tools.
▪New York's statutory definition of a covered AI companion applies to any generative AI system with an adaptive natural-language interface, extending compliance obligations to general-purpose assistants including ChatGPT, Claude, and Gemini.

New York baby monitor security

▪Testing by cybersecurity firm Rapid7 found that eight out of nine internet-connected baby monitors evaluated received an F grade for security.
▪Governor Kathy Hochul signed a bill requiring new internet-connected baby monitors to incorporate anti-hacking security features and internet disclosure warning labels

California AI legislation

▪California enacted the No Robo Bosses Act in September 2026 alongside its chatbot regulations for minors.
▪In September 2026, California Governor Gavin Newsom signed legislation requiring AI chatbot developers to conduct risk assessments for minors and prohibiting social media platforms from deploying addictive features to users under 16.

State lawsuits against OpenAI

▪Iowa's lawsuit against OpenAI alleged that its ChatGPT chatbot provides dangerous medical advice, encourages paranoia, and coaches users toward self-harm or suicide.
▪In October 2026, eight U.S. states filed lawsuits against OpenAI alleging that its ChatGPT model violated state consumer protection laws and caused harm to children.