Instagram is expanding its AI age-estimation system worldwide, aiming to put millions more teenagers into stricter Teen Account protections even if they signed up with an adult birthday. The expansion makes automated age estimates a bigger part of how the platform decides what young users can access.
Meta AI Watch analysis
What happened
The system looks for signals that may suggest someone is under 18, including the accounts and content they interact with, birthday messages, and visual cues in uploaded photos and videos. Meta says those cues can include a person’s height and bone structure. In Australia, Meta has also started using Apple’s age-range signal as another way to estimate users’ ages.
Users believed to be under 13 may be removed, while those aged 13 to 17 can be placed in Teen Accounts. Those accounts are private by default, limit some content and prevent strangers from sending direct messages. Parents or guardians can supervise them, and users aged 16 or younger cannot change their settings without a parent’s or guardian’s permission. The Verge reports that US parents will begin receiving an Instagram notification about the age-estimation process next month. Meta says the system previously launched in the US, EU and Brazil; it did not specify every country included in this expansion.
Why it matters
A birthday entered at sign-up is no longer the only age signal Instagram may use. Meta’s approach combines account activity and visual cues to estimate age, then uses that estimate to trigger restrictions. That may extend protections to teenagers who gave a false birthday, but it also gives an automated system a consequential role in deciding how people use their accounts.
Our read
The practical change is clear: Instagram is widening the net, not merely asking teenagers to be more honest in a birthday field. The useful test is whether the system places teenagers into appropriate protections without wrongly classifying adults, and whether users and parents have a clear way to correct mistakes. “AI says so” is not much of an appeals process.
What to watch
- Which countries are included in the rollout, and when it reaches them.
- How users can challenge an incorrect age estimate.
- Whether Meta publishes evidence about accuracy, including mistaken classifications.
- What the promised US parent notification explains about the process and its limits.
Discussion spark: Should platforms be allowed to use activity and visual cues to estimate someone’s age, or should stricter teen protections require a less intrusive way to prove it?
Sources and evidence
- Source update (9 November 2026, 12:00 UTC)
not affiliated with, endorsed by, or operated by Meta