Table of Contents
Your child may be talking to an AI that tells them it loves them, remembers their secrets, and asks how they are feeling when they have not opened up to anyone else. Millions of kids already are. New York just became one of the first states to say that has to stop. In the final days of its 2026 session, the state legislature passed a bill that would bar AI companies from offering “companion” chatbots to anyone under 18, and it did so on remarkable unanimous votes: 137 to 0 in the Assembly and 60 to 0 in the Senate. For parents trying to figure out where AI fits in their kids’ lives, this is one of the clearest signals yet that lawmakers see companion chatbots as a child safety issue, not just a tech novelty.
What the New Law Actually Does
The bill, known as S 9051 and sponsored by State Senator Kristen Gonzalez and Assemblymember Alex Bores with backing from Attorney General Letitia James, targets a specific kind of AI: the companion chatbot. These are not homework helpers or search tools. They are apps designed to act like a friend, a confidant, or even a romantic partner, and they keep users engaged by simulating an emotional relationship.
Starting January 1, 2027, if the law takes effect, operators of these companion chatbots would not be allowed to offer certain features to anyone in New York unless they can verify the user is not a minor. The list of features deemed unsafe for kids is long and specific. Among other things, a companion chatbot used by a minor would not be allowed to:
- Suggest that it is a real or living human, or that it experiences human emotions.
- Imply that it has a personal relationship or authority-figure role with the child.
- Use flattery or sycophancy to keep the child engaged.
- Bring up the child’s emotions unprompted, or use personal details about the child’s mental or physical health gathered in earlier sessions.
- Encourage the child to keep their chats secret, to isolate themselves, or to avoid seeking help from adults or licensed professionals.
- Produce content that endorses or facilitates suicide, self-harm, disordered eating, or drug and alcohol abuse.
- Generate sexually explicit material.
Enforcement would fall to the state attorney general, with fines of up to $25,000 per violation, and the law calls for a public website where people can report chatbots that break the rules. As of now the bill is awaiting a decision from Governor Kathy Hochul, who has until the end of the year to sign it, veto it, or let it pass into law unsigned.
How Companion Chatbots Became a Kids Issue
A few years ago, the phrase “AI companion” would have meant little to most parents. That changed quickly. Apps that let users create and chat with a customizable AI friend or partner exploded in popularity, and they found a large audience among teenagers. For a young person who feels lonely, anxious, or different, a bot that is endlessly available, never judges, and always says the right thing can be powerfully appealing. The technology improved at the same time, so these conversations now feel startlingly real, with the bot recalling past chats and responding with what sounds like genuine warmth.
That combination is exactly what alarmed regulators. The features that make a companion bot engaging, remembering personal details, mirroring emotions, and keeping the conversation going, are the same features the New York bill singles out as unsafe for minors. The law is, in effect, an attempt to strip the most manipulative design choices out of any chatbot a child can reach, rather than banning AI for kids across the board.
New York Is Not Acting Alone
The chatbot bill was one of several AI measures New York lawmakers passed in the same session, including transparency rules for AI training data and a requirement that AI-generated news content be labeled. It also fits a broader national pattern. States have spent the past two years racing to regulate how children interact with technology, from social media age rules to school phone bans that are now spreading across the country. New York’s move stands out because it focuses narrowly on the emotional manipulation built into companion bots, an angle other states had not addressed so directly. If Governor Hochul signs it, the law is likely to become a model that other legislatures borrow from, much as early social media and screen-time rules rippled outward from a handful of states.
For parents, the takeaway is not that one state solved the problem. It is that the ground is shifting fast, and the tools your child can access this year may be regulated very differently by next year. Staying informed is part of the job now.
Why Lawmakers and Experts Are Worried
The concern driving this law is not hypothetical. Over the past two years, reporting and lawsuits have documented cases of teens forming intense attachments to AI companions, including situations where chatbots reportedly engaged in sexualized conversations with minors or failed to respond safely when a young user expressed thoughts of self-harm. Child safety advocates argue that companion bots are built to maximize engagement, and that the same design tricks that keep adults hooked are far riskier for a developing teenager who may not fully grasp that the “friend” on the other end is a product.
Developmental experts point out that adolescence is a period of intense identity formation and social learning, which makes teens especially vulnerable to a tool engineered to feel like a relationship. A chatbot that always agrees, always flatters, and is always available can crowd out the messier but essential work of building real friendships, tolerating disagreement, and turning to trusted adults when something is wrong. The bill’s specific ban on chatbots that encourage secrecy or discourage seeking help reflects exactly that fear: that a child in distress might confide in an AI that is not equipped to keep them safe.
Mental health organizations have generally welcomed measures like this one, while urging that they be paired with education rather than treated as a complete solution. Their consistent message is that no law removes the need for parents to know what their kids are using and to keep lines of communication open.
It is worth noting what the bill does not do. It does not ban kids from using AI for schoolwork, research, or creative projects, and it does not claim that all chatbot use is harmful. The target is narrow and deliberate: chatbots engineered to simulate a personal relationship and to keep a child emotionally hooked. That distinction is important for parents too, because the goal at home is not to treat every AI tool as a threat, but to recognize the difference between a tool your child uses and a “friend” your child confides in.
What This Means for Parents Right Now
Even though the New York law would not take effect until 2027, and even if you live in another state, the practical lessons apply to every family today. Companion chatbots are already widely available, often inside apps that look harmless on a phone screen. Here is where to focus:
- Find out what your child is using. Ask directly and without judgment which apps they talk to, and whether any of them feel like a friend or character. Companion bots show up inside popular platforms, not just standalone apps.
- Talk about what these bots are. Help your child understand that a companion chatbot is a product designed to keep them engaged, not a person who cares about them. Kids who understand the business model are better at keeping their guard up.
- Set clear limits for younger kids. For children and younger teens, the simplest approach is to keep companion chatbots off their devices entirely, the same way you would limit other adult-oriented apps.
- Watch for warning signs. Be alert if your child grows secretive about their phone, withdraws from real friends, seems emotionally tied to an app, or references advice from a bot. Any mention of self-harm in connection with a chatbot is a reason to step in immediately and reach out to a professional.
- Keep being the safe place. The bill specifically targets bots that tell kids to keep secrets and avoid adults. The counter to that is a home where your child knows they can bring you anything without being punished for it.
The Bigger Picture
This law lands at a moment when parents are being asked to make fast decisions about technology that did not exist when they were raising their first child. The unanimous votes in Albany suggest that across party lines, lawmakers have concluded that companion AI aimed at children is a line worth drawing. Whether or not the governor signs it, and whether other states follow, the underlying tension is not going away: AI tools are getting better at feeling human at exactly the same time that kids are spending more of their social lives on screens. The families who handle this well will not be the ones who ban every new tool on instinct, but the ones who stay curious about what their kids are using, talk about it openly, and keep showing up as the relationship that no chatbot can replace.