Meta's age-verification plan using AI raises privacy concerns

Meta's $18 billion settlement with states requires the company to improve age assurance to identify child users. The company plans to use AI to estimate ages based on user behavior, but this raises privacy concerns for both minors and adults. Meta has a year to refine its approach, which could become industry standards.

EXPANDED:

Meta's AI age-estimation system analyzes behavioral signals such as posting times, friend networks, and engagement patterns, alongside visual cues from photos like height or bone structure. This method estimates age rather than verifying birth dates, and the settlement mandates regular third-party audits with specific targets for false positive rates—cases where minors are mistakenly classified as adults. If a user is removed for being under 13, Meta must also check the ages of their connected friends.

Other platforms employ similar tactics. Google's YouTube uses AI based on watch history, while identity firms like Yoti and Persona process video selfies for age estimates, a method Roblox