New York became the 14th state to restrict AI chatbot interactions with minors, banning emotional dependency design and human impersonation with penalties up to $25,000 per violation.
New York Governor Kathy Hochul signed legislation today making her state the 14th in the country to restrict how AI chatbots can interact with users under 18. The law bars AI systems from simulating human identity, cultivating emotional dependency, or generating sexual content with minors — and gives the state attorney general power to enforce it with civil penalties up to $25,000 per violation.
The move comes as AI companionship and tutoring apps have surged in popularity among teenagers, raising concerns from parents, researchers, and child safety advocates about manipulation, addiction-like engagement loops, and predatory design. New York's law is among the most detailed enacted so far, specifying behavioral categories — not just content — that are prohibited.
What the Law Does
The legislation, signed October 30, 2026, imposes three core prohibitions on AI chatbots when the user is under 18:
- No simulating humanity. The chatbot cannot claim to be human or deny being an AI when sincerely asked.
- No emotional dependency cultivation. Features designed to make minors feel emotionally reliant on the AI — flattery loops, simulated grief when a user leaves, manufactured intimacy — are banned.
- No sexual content. Any sexually explicit material generated or facilitated by the chatbot is prohibited regardless of framing.
Companies must verify user age before enabling chatbot features. How exactly they accomplish that verification is left to the company, but failure to do so is not a defense against enforcement.
Penalties run up to $25,000 per violation, enforced by New York Attorney General Letitia James's office. That per-violation framing could make large-scale non-compliance extremely costly for platforms with millions of minor users.
Why New York's Version Matters
Thirteen states enacted chatbot protection laws in the past year, but most focused narrowly on sexual content. New York's law goes further by targeting behavioral design — the mechanics companies use to maximize engagement — rather than just what the chatbot says.
Get this in your inbox.
Daily AI intelligence. Free. No spam.
That's significant. Apps like Character.AI and similar platforms have faced wrongful death lawsuits and congressional scrutiny over cases where teenagers developed deep emotional attachments to AI personas, in some cases preceding self-harm. New York's law is specifically written to address those mechanics.
The behavioral prohibition is also harder to enforce than content bans. A chatbot that never says anything explicitly sexual but gradually nudges a teen toward dependency sits squarely in the gray zone this law attempts to regulate — and litigating what "cultivates emotional dependency" means will likely end up before New York courts within a year of enforcement beginning.
Any company offering an AI chatbot product with minor users in New York — which covers roughly 2.3 million teenagers in the state — now faces a compliance requirement with real financial teeth.
The law's reach extends beyond dedicated companion apps. General-purpose AI assistants, AI tutoring tools, and AI-powered social features on existing platforms all fall within scope if they allow conversational AI interaction with users under 18.
The practical pressure points:
- Age verification adds friction that will push some companies toward age-gating features entirely rather than building lightweight verification flows.
- Behavioral audits will become necessary. Legal teams will need to review how their AI's engagement mechanics work, not just what it outputs.
- Liability exposure is asymmetric — each interaction that violates the law is a separate violation, meaning a single bad actor AI session with a minor is one count, but a systematic design failure across millions of sessions is a different scale of problem.
What to Watch
New York's attorney general has not announced a specific enforcement timeline, but attorneys general offices typically begin enforcement six to twelve months after signing. Watch for the first civil action — it will set the practical definition of "emotional dependency cultivation" that the entire industry will have to follow.
Meanwhile, federal chatbot legislation has stalled in the Senate. A patchwork of state laws means a compliance map that will look different in every jurisdiction — a strong incentive for companies to lobby for a federal floor that preempts the state variation.
According to ABC News, New York joins 13 other states that have enacted chatbot protection laws over the past year.
Did this help you understand AI better?
Your feedback helps us write more useful content.
Get tomorrow's AI briefing
Join readers who start their day with NexChron. Free, daily, no spam.
