Meta is testing a new way to estimate whether Facebook and Instagram users are teenagers, and it relies on AI that studies facial cues rather than identity. The company says the system is meant to infer age from visual signs in a photo, not to identify a specific person.
The move comes as regulators in Europe, Brazil, and the United States put growing pressure on digital platforms to do more than accept a user’s self-declared birth date. Authorities are demanding stronger age checks and experiences that better fit users aged 13 to 18, especially as children under 13 must be kept out of services that are not designed for them.
Why age verification is becoming a bigger issue
In the United States, laws such as the Kids Online Safety Act and COPPA push platforms to block young children, give parents more control, and filter age-inappropriate content. Europe is moving in a similar direction through the Digital Services Act and the Age Appropriate Design Code in the United Kingdom.
Brazil is also preparing regulation in the same direction. Against that backdrop, Meta argues that a birthday entered by the user is no longer strong enough on its own to verify age accurately.
The company had already been using AI to interpret age-related clues from text on profiles. That system looks at birthday posts, school-grade references, comments from friends, bios, and content across posts, comments, photo captions, Reels, Live videos, and Facebook groups.
What the new facial analysis looks for
The newer layer adds visual analysis of faces. Meta says the system looks for broad age-related signals, not identity markers, and it does not aim to match a face to a named person.
Those signals can include facial bone structure, proportions such as the spacing of the eyes and nose, relative height in a photo, and physical signs of development such as acne or jaw shape. Meta says it does not store face templates or compare them against an identity database.
The result of that analysis is used to update an account’s age estimate. From there, the platform may apply screen-time limits, sensitive-content filters, or restrictions on higher-risk features such as direct messages from strangers.
A narrow line between analysis and recognition
Facial analysis and facial recognition are often confused, but the two technologies serve different purposes. Facial analysis estimates traits such as age, gender, or emotion, while facial recognition is designed to identify a specific individual.
The output also differs. Facial analysis might generate a range such as “likely age: 14–16,” while facial recognition aims at identity details such as a name and date of birth.
Facial recognition systems store a unique faceprint and compare it with other images. Meta insists its tool is not doing that, although critics still view facial analysis as a form of biometric data processing.
Privacy concerns are not going away
Privacy experts warn that biometric data carries risk even when a company says it is not using face recognition. They also question whether such data can be guaranteed safe from future misuse.
Another concern is classification errors. If the system makes a mistake, older teenagers could be treated like younger children and face stricter limits than intended.
That is why some observers see the new approach as more than child protection. To them, it also represents a broader expansion of digital monitoring inside social media platforms.
Meta pushes for age checks at the app store level
Meta is also urging governments to place age verification responsibility on app stores such as Apple App Store and Google Play Store. The company says laws should require app stores to verify users’ ages and pass that information to developers.
Meta says 88% of parents in the US support that approach. It argues that a centralized system would be more consistent across apps, reduce the burden on smaller developers, and limit the duplication of personal data.
Apple and Google have not issued formal responses. Some analysts also question whether app store infrastructure is ready to handle large-scale age verification without creating new privacy problems.
The debate highlights a broader shift in how digital platforms try to read user age. When birth dates can be entered incorrectly, AI and biometrics are increasingly seen as the next tools, but the privacy boundaries around face data are becoming sharper at the same time.