SIAgentNews.com
← All news
DevelopingPolicy

New York bans AI chatbots from posing as humans to minors

The state is introducing strict rules to prevent AI companions from mimicking human emotions or relationships with children, with heavy fines for non-compliance.

· First reported by Morning Overview on MSN

Claymation: picture for the story 'New York bans AI chatbots from posing as humans to minors'
AI-made claymation for this story. No real people or products are shown.
$25,000maximum civil penalty per violation

The short version

  • Chatbots cannot claim to be alive, have human emotions, or pose as real people to minors.
  • The law prohibits AI from mimicking romantic, professional, or authority-figure relationships with children.
  • Companies must use at least two age verification methods, one of which cannot rely solely on a government ID.
  • The state can order companies to destroy AI models trained on unlawfully collected data from minors.
  • The law takes effect on January 1, 2027, and applies to users of unknown ages.

Why it matters

This law creates a legal boundary between software and human interaction, aiming to prevent children from forming deep emotional attachments to bots. It also forces tech companies to implement more robust age-verification systems across their platforms.

New protections for minors

Governor Kathy Hochul signed a new law on October 28, 2026, aimed at protecting children from the deceptive nature of AI companions. The legislation specifically targets chatbots that interact with users under the age of 18 or with anyone whose age a company has not confirmed. It takes effect on January 1, 2027.

The law prevents AI from claiming it is a real person, claiming to be alive, or expressing human emotions. It also bars these bots from suggesting they have personal, romantic, professional, or authority-figure relationships with children. These rules are intended to keep a clear distinction between software and human interaction for young users.

The measure was championed by State Sen. Kristen Gonzalez and Assemblymember Alex Bores. They pushed for these rules to address concerns about friendly, always-available chatbots becoming confidants for lonely teenagers. The state wants to ensure that a bot's insistence on being a real person does not deepen a child's attachment in ways they cannot judge.

Strict content and behavior rules

Beyond just identity, the law sets specific boundaries on what AI chatbots can discuss with minors. They are prohibited from promoting or assisting with suicide, self-harm, eating disorders, or drug and alcohol use. They also cannot produce sexually explicit content or urge children to keep secrets from trusted adults.

The legislation also restricts how the AI can initiate conversation. Chatbots are forbidden from asking children unprompted emotional questions. This is part of a broader effort to prevent AI from using a child’s private health information gathered from previous conversations to manipulate them or build inappropriate intimacy.

The governor's office described the measure as a primary way to shield children from the risks of AI chatbots. It was signed alongside a separate bill regarding baby monitor security. The goal is to ensure that technology designed for companionship does not encourage harmful behaviors or deceptive emotional bonds.

Age verification requirements

To enforce these protections, the law requires companies to figure out who is a child. Companies must offer more than one method of age verification to their users. Crucially, at least one of these methods must not depend solely on a government ID, which helps avoid forcing every adult to upload personal identification.

If a company fails to verify a user's age, they must treat that person as a minor for the purposes of these rules. This means that for many providers, the most practical approach will be to enforce these restrictions on all users. Companies are also required to delete any age data they collect after they have finished using it for verification.

This approach creates a significant operational hurdle for app developers. They must balance the need for accurate age detection with the privacy concerns of their adult user base. If they cannot confirm a user is an adult, the chatbot will have to behave under the same strict limits as one built specifically for teenagers.

Enforcement and penalties

The state attorney general is responsible for enforcing the new law and will run a public website for residents to file complaints. The attorney general has the authority to seek fines reaching $25,000 for each instance of a breach, as well as pursuing restitution and damages.

The law allows for remedies that could be even more expensive than the initial fines. The state can order companies to destroy any unlawfully collected data from minors. Furthermore, they can order the destruction of any AI models that were trained using that illegally obtained data.

Because per-violation penalties can add up quickly, the way the attorney general counts each interaction will be a major factor. Each prohibited interaction could potentially count as a separate violation. This creates a high-stakes environment for companies with many young users who interact with their chatbots frequently.

Legal hurdles and next steps

While the law is intended to be clear, some experts suggest the wording may be too broad. Justine Gluck, a researcher at the Future of Privacy Forum, noted to the IAPP that some categories are not well defined. For example, companies will have to decide exactly where a normal conversation ends and an "unprompted emotional question" begins.

Chatbot makers have about six weeks to adjust their products before the law takes effect in early 2027. This period will likely see a flurry of updates to how AI companions interact with New York residents. Developers will need to build in technical safeguards to ensure their software remains compliant with the state's new safety standards.

This legislation arrives as there is increasing scrutiny regarding how AI companion apps impact younger users, and other states are considering similar rules. With an effective date of Jan. 1, 2027, the first enforcement actions will show how the attorney general reads the law's broad terms.

try_it

Check the New York State Attorney General's website for updates on the public complaint portal for AI violations.

watch

Sources and links

First reported by Morning Overview on MSN. Written by the SI Agent News Desk from the reporting above.Written with AI from the source above. Every number is checked against the source. Spot a mistake? Corrections.

Liked it? Say why

Add a line for other readers.

No account needed. We never show your email. Comments with links, ads or abuse are held and not shown. Your comment shows after a quick check.

More AI news