New York Bans ‘Unsafe’ Kids’ Chatbots as States Test the Limits of AI Regulation

New York has enacted a first‑of‑its‑kind state law that bans “unsafe” behavior by AI chatbots when they interact with children. Reported on October 28, the law goes beyond simple disclaimers or content warnings and directly targets how youth‑facing chatbot systems are designed and operated.

Under the statute, companies offering chatbots to minors must build in safeguards to block content that promotes self‑harm, sexual material, or other harmful behavior, and are expected to implement age‑appropriate modes and some form of age checks. The focus is squarely on the underlying algorithms: providers are pushed to prove that their models are tuned and filtered to avoid dangerous responses, not just that they publish safety guidelines.

Policy analysts say New York’s move could become a template for other U.S. states as lawmakers search for concrete ways to curb AI risks for children, even while broader national rules remain stalled. But a growing patchwork of state‑level obligations may also raise compliance costs and technical complexity for AI firms, which must adapt one set of models to diverging local safety standards across the country.