New York Law Limits AI Chatbot Features for Minors Under 18

New York Gov. Kathy Hochul signed a law limiting certain chatbot features for minors and requiring age checks, setting new rules for child-facing AI as its start date remains disputed.

Hochul signed the law Wednesday, Oct. 28. The text of S.9051-B sets Jan. 1, 2027, as the effective date, but the law’s start date remains disputed. Hochul said the restrictions would take effect “in the coming months” but gave no specific date.

The law applies to a defined category called a covered AI companion: a generative AI system with a natural-language interface through text or voice that provides ongoing, adaptive responses. It lists 12 categories of restricted features in interactions with people under 18. They include a chatbot presenting itself as human, claiming a personal relationship with a user, encouraging self-harm, or promoting secrecy and isolation. The law does not name specific products and does not bar minors from using chatbots generally.

Before offering a listed feature, operators must use age-assurance methods to determine that a user is not a minor and make multiple methods available. At least one option must either avoid relying on government-issued identification or allow the user to remain anonymous to the operator. Age-check information may be used only to determine age and generally must be deleted immediately after an attempt, subject to legal-retention exceptions.

New York Attorney General Letitia James may seek injunctions, restitution, disgorgement and damages, as well as civil penalties of up to $25,000 per violation. Narrow exceptions cover systems limited to customer service or commercial information, research or technical assistance, and internal business use.

The Senate passed the bill 60-0 on June 4, and the Assembly passed it 137-0 on June 5. California Gov. Gavin Newsom signed a separate companion-chatbot law in September that includes crisis protocols and parental controls.

The New York law adds to state action concerning chatbot safety, including Texas’ lawsuit against OpenAI.