Meta is expanding the AI age estimation system it uses on Instagram to every market, aiming to catch teenagers who did not give their correct age.
According to Meta's announcement, the rollout will automatically place "millions more teens on Instagram into Teen Account protections." The company first launched the AI-powered age assurance updates in the US, EU, and Brazil in 2025.
The system works by looking for signals that likely indicate whether someone is under 18. Meta listed the other profiles and types of content an account interacts with, the birthday messages it receives, and "general themes and visual cues" from uploaded photos. These checks supplement the birthday a user typed in at sign-up rather than standing in for it.
The approach responds to an old weakness. A self-reported birthday is only as honest as the person entering it, and a 13-year-old can type a different year in seconds. Behavior is harder to fake consistently, and a stream of "happy 15th" messages is a fairly blunt tell.
But inference is a guess, not a verification. The description of the system gives no error rates, so it is unclear how often adults get flagged as teens or teens slip through. Those two failures carry different costs: the first restricts people who did nothing wrong, the second leaves the protections unused. The photo signal also invites a question about what Meta's models are trained to recognize, which the announcement language does not answer.
The larger point is that Meta is treating age as something to be detected rather than declared. Whether that satisfies regulators and parents will depend on numbers Meta has not yet put in front of the public.