The Youth AI Privacy Act protects minors by prohibiting AI chatbot developers from collecting their personal data for profiling, training, or advertising, while mandating strict design safeguards and transparency requirements.
Edward "Ed" Markey
Senator
MA
The Youth AI Privacy Act protects minors by prohibiting AI chatbot developers and deployers from collecting, profiling, or using the personal data of users under 18 for commercial purposes or algorithm training. The bill mandates clear disclosures about the non-human nature of AI, restricts addictive design features, and limits data retention to protect young users. Enforcement is overseen by the Federal Trade Commission and state attorneys general, with additional provisions for parental legal action and expanded federal research into the impact of AI on youth development.
Youth AI Privacy Act Bans Targeted Ads and 'Addictive' Features for Minors on Chatbots
The Youth AI Privacy Act is a direct response to the growing trend of teenagers using AI chatbots for everything from homework help to emotional support. At its core, the bill prohibits AI companies from collecting, profiling, or selling the personal data of users under 18 unless that data is absolutely necessary for the app to function. It essentially draws a line in the sand: if a company knows a user is a minor, they can no longer use that kid’s private conversations to train their algorithms or serve them targeted advertisements. The goal is to shift AI from a data-mining tool into a safer, more transparent utility for the next generation.
No More 'Ghost' Interactions
One of the biggest changes involves how these AI tools present themselves. Under Section 4, companies must clearly disclose that the user is talking to a bot, not a human, at the start of every session and every 30 minutes thereafter. This is designed to break the 'illusion' of human connection that can lead to emotional dependency. For a parent whose middle-schooler is staying up late venting to a chatbot, this means the app will regularly tap the kid on the shoulder to remind them they’re talking to code. Furthermore, the bill bans 'compulsive' design features like usage streaks, badges, and push notifications that are currently used to keep users hooked. It’s an attempt to treat AI more like a library book and less like a slot machine.
Privacy by Default
When it comes to the 'fine print' of data, the bill gets technical but tough. Section 5 stops companies from using a minor’s past data to personalize responses unless it was gathered in the current session. This prevents the AI from building a long-term 'psychological profile' of a child over several years. For developers, this is a significant hurdle; they’ll have to re-engineer how their models learn and interact. While the bill doesn't force companies to implement invasive ID checks (age gating), it uses a 'knowledge' standard—meaning if it’s obvious based on the data that a user is a kid, the company is legally on the hook to protect them.
Enforcement and the 'Parental' Safety Valve
This isn't just a list of suggestions; it has teeth. The Federal Trade Commission (FTC) and state attorneys general are tapped to lead the charge on enforcement, treating violations like deceptive business practices. However, the real-world kicker is the 'Private Right of Action' in Section 9. This allows parents or legal guardians to personally sue AI developers if their child’s privacy rights are violated. While this provides a powerful tool for families to hold tech giants accountable, it also means smaller AI startups will need to be extremely careful with their compliance costs to avoid potentially bankrupting lawsuits. The bill also sets aside $50 million a year for research into how this tech actually affects kids' brains, ensuring that future laws are based on hard data rather than just guesswork.