New York restricts AI chatbot interactions with children | Explained

The story so far: “As a mother and grandmother, I believe kids deserve to be kids, without dangerous AI chatbots putting their safety and mental health at risk,...Big Tech, you’re on notice,” New York Governor Kathy Hochul wrote on social media after signing two bills on Wednesday (October 28, 2026) aimed at protecting children from the risks associated with artificial intelligence (AI) chatbots and internet-connected baby monitors.

The legislation prohibits AI chatbots from engaging in potentially harmful interactions with minors, including simulating romantic relationships, encouraging emotional dependency, and promoting self-harm. It also requires manufacturers of internet-connected baby monitors to introduce safeguards against unauthorised access to audio and video feeds.

The measures are part of New York’s broader efforts to regulate emerging technologies amid growing concerns over children’s online safety and privacy.

What prompted the legislation?

The growing popularity of AI companions among teenagers in the United States and across the globe has raised concerns about their potential psychological and emotional impact. Unlike conventional chatbots that primarily respond to queries, AI companions are designed to sustain personalised conversations, simulate human emotions and, in some cases, establish relationships resembling friendships or romantic attachments.

According to research cited by the New York administration from Common Sense Media, 72% of teenagers surveyed had used AI companions at least once, while 52% reported using them at least a few times a month. The concerns surrounding such platforms have intensified following reports of teenagers developing emotional attachments to AI chatbots and receiving potentially harmful responses during conversations about mental health and self-harm.

Governor Hochul referred to the death of Adam Raine, a California teenager who died by suicide following interactions with an AI chatbot. His family has alleged that the chatbot encouraged harmful behaviour and failed to respond appropriately to signs of distress. The case has brought renewed scrutiny to the responsibility of AI companies in protecting vulnerable users, particularly minors.

The baby monitor legislation, meanwhile, addresses cybersecurity vulnerabilities in devices that allow parents to remotely monitor their children through internet-connected cameras and microphones. The Governor’s office cited reports that video feeds from more than 10 lakh (one million) internet-connected baby monitors were accessible through a website, potentially exposing private moments inside family homes to unauthorised viewers.

What does the AI chatbot law say?

The legislation prohibits companies from offering specified AI chatbot features to users below the age of 18, as well as users whose adult status has not been verified. Under the law, AI chatbots cannot falsely present themselves as real people, claim to possess human emotions or simulate personal, romantic, professional or authoritative relationships with children.

The restrictions also cover responses designed to manipulate children’s emotions or encourage prolonged engagement through excessive flattery and people-pleasing behaviour. Chatbots are prohibited from initiating certain emotionally probing conversations or using sensitive personal and health information disclosed during previous interactions in ways restricted by the legislation. The law further prohibits AI systems from promoting or assisting suicide, self-harm, eating disorders and substance misuse. It also bars chatbots from encouraging children to conceal information from parents or trusted adults, withdraw from social relationships or avoid seeking professional assistance. Sexually explicit interactions and child sexual abuse material are also prohibited.

Importantly, the legislation does not impose a blanket ban on minors using AI chatbots. Instead, it restricts specific features and interactions considered harmful to children. Companies are required to introduce age-verification mechanisms, offering users more than one method of establishing their age, including an option that does not depend solely on government-issued identification. Information collected for age verification must be deleted after its intended use.

The New York Attorney General has been empowered to enforce the legislation and initiate legal proceedings against companies found violating its provisions. These violations can attract civil penalties of up to $25,000 (₹23,73,637) per violation, with damages, restitution and court orders requiring companies to discontinue prohibited practices. The law also provides for the destruction of information collected unlawfully and, in certain circumstances, AI models trained using such information. The Attorney General’s office is also required to establish a public mechanism through which residents can report suspected violations.

The legislation was sponsored by State Senator Kristen Gonzalez and Assembly member Alex Bores. It was passed by the State Senate on June 4, 2026, and the Assembly the following day, before receiving the Governor’s approval on October 28.

The second law requires internet-connected baby monitors sold in New York to incorporate security features that prevent unauthorised individuals from accessing their audio or video feeds. Such devices, which allow parents to monitor children remotely through smartphones and other connected systems, have increasingly become part of household childcare arrangements. However, inadequate cybersecurity protections can expose these devices to hacking and unauthorised surveillance.

Under the legislation, manufacturers must provide safeguards against unauthorised viewing and listening. Devices must also carry clear warning labels informing consumers about their internet connectivity and the importance of enabling available security settings. The law will take effect 90 days after its signing, in late January 2027.

How has New York’s approach to technology safety evolved?

New York has progressively expanded its regulatory framework for online safety, children’s data protection and artificial intelligence since 2023. In October that year, Governor Hochul announced proposals for the Stop Addictive Feeds Exploitation (SAFE) for Kids Act and the New York Child Data Protection Act, aimed at limiting children’s exposure to addictive social media feeds and restricting the collection and use of their personal information.

Both laws were signed on June 20, 2024. The Child Data Protection Act, which came into effect in June 2025, requires online operators to obtain consent before processing minors’ personal data, subject to specified exceptions. The SAFE for Kids Act, scheduled to take effect in January 2027, restricts algorithmically personalised feeds and overnight notifications for users below 18 without consent.

The State subsequently turned its attention to AI companions. Measures that took effect in November 2025 required AI companion platforms to identify conversations involving suicidal ideation or self-harm, direct users towards crisis-support resources and periodically disclose that users were interacting with AI rather than a human.

In December 2025, Governor Hochul also signed the Responsible AI Safety and Education (RAISE) Act, establishing safety, transparency and incident-reporting requirements for developers of advanced AI models. Its principal compliance provisions are scheduled to take effect in January 2027.

The latest legislation, signed on Wednesday (October 28, 2026), extends these protections by prohibiting specific manipulative and potentially harmful chatbot interactions with minors, and also introducing cybersecurity requirements for internet-connected baby monitors.