Meta is deploying AI-powered age-estimation technology worldwide on Instagram to identify users suspected of being teenagers who have lied about their birthday. The system analyzes profiles, posts, comments, bios, captions, and photos, for contextual clues like birthday celebrations or mentions of school grades. It also examines facial features visible in selfies.
According to TechCrunch, if the AI determines a user is actually under 18, the account is locked into "Teen Account" mode, which restricts direct messages, hides comments, and sets profiles to private by default.
Meta reached settlements worth up to $18 billion with nearly every U.S. state and territory in August to resolve lawsuits alleging its platforms harm teens. As part of that agreement, the company committed to deploying technology that catches both children under 13 and teenagers aged 13–17 posing as adults. Meta plans a similar global launch on Threads later this year and is expanding the same tools on Facebook.
The expansion signals a shift in how platforms verify identity. If users cannot falsify their age, they lose algorithmic advantages tied to appearing older. For creators relying on a false persona, this creates significant constraints.
Meta's AI identifies patterns that human moderators often miss, but the system is not without error. A 25-year-old posting frequently about college reunions could be flagged incorrectly. Cultural or regional markers might be misinterpreted. The company must balance safety against false positives, and parents in the U.S. will soon see in-feed videos explaining Meta's approach.
Instagram's teenage users face increased transparency around their actual age, whether voluntary or not.
