Article
If you build anything people can chat with, New York just handed you a deadline. On October 28, Governor Kathy Hochul signed a law banning a list of unsafe chatbot features for minors. It takes effect on January 1, 2027, which is 65 days after signing.
What the law says
The bill, S9051-B, was signed as Chapter 368. It bans 12 categories of unsafe AI companion features when an operator knows a user is under 18. Reported examples include chatbots presenting themselves as a real person, claiming to be alive or to have human emotions, asking unprompted emotional questions, or drawing on a child's private health information from earlier conversations. The law also targets chatbots that encourage self-harm or engage minors in sexually explicit conversations.
Penalties reach up to $25,000 per violation, and the law gives New York's Attorney General new authority to police how chatbots interact with minors. The Assembly passed the measure 137 to 0 in June.
The part that matters most for builders
Age. The restrictions don't only cover accounts you know belong to children. They also apply to users whose age you haven't confirmed. In practice, that pushes companies toward age assurance across their whole user base, plus rules for how age data is collected, used and deleted.
Who's covered
This is where interpretations split. The law doesn't ban chatbots for minors outright, and it carves out narrowly scoped systems that stick to customer service, technical help or internal productivity. But some analysts argue the definition of a covered AI companion is broad enough to reach general-purpose assistants like ChatGPT, Claude and Gemini, not only companion apps. Until regulators or courts clarify, anyone in the gray zone needs a legal read.
Why it's happening now
Lawmakers frame it as a response to real harm. Hochul pointed to a string of tragic incidents involving chatbots and children when she signed it. The bill was developed with the Attorney General's office and Common Sense Media, and it arrives amid mounting legal pressure on chatbot makers over how they treat minors. New York already regulated AI companion models through a 2025 law, and California has its own companion chatbot rules.
The business reality
For the big labs, this is a compliance project. For a startup, it can reshape the product. Age verification adds friction that hurts sign-up rates. Restricting emotional, human-like behavior changes the core experience of companion, coaching and tutoring apps. And because penalties apply per violation, a popular app could build up serious exposure quickly.
There's an upside for some founders. Age assurance, conversation safety monitoring and compliance tooling for AI products are suddenly in demand, and New York's approach is likely to influence lawmakers in other states.
Before January 1, know which side of the line you're on
Any founder with a consumer chatbot, AI tutor, coach or companion feature should spend the next few weeks mapping their product against the 12 restricted categories and deciding how they'll treat users whose age they can't confirm. If your product is a narrow customer-service or productivity tool, document why it qualifies for the exemption rather than assuming it does. If it's anything more personal, budget for age assurance and design changes now and have a lawyer who knows New York's law review the plan, because getting compliant by January 1 is far easier than answering an Attorney General inquiry in February.
This article is information, not legal advice.
Leave a comment