Meta Platforms Inc. has announced the deployment of an upgraded, artificial intelligence-powered age assurance framework across all its primary applications—Instagram, Facebook, and Messenger—within Pakistan. The initiative focuses on accurately identifying user demographics to automatically enforce default safety experiences, such as restricted “Teen Accounts,” for users under the age of 18.
The newly introduced mechanism relies on two primary AI layers to cross-verify user ages without relying entirely on traditional identity documentation. The first layer performs behavioral and contextual text processing, scanning user profiles, post captions, comments, and bios for contextual markers. The system looks for explicit age indicators, such as mentions of school grades, specific academic milestones, or birthday celebrations. This text-based tracking operates proactively across multiple product features, including standard feeds, Instagram Reels, Instagram Live broadcasts, and Facebook Groups.
The second, more technically complex layer involves the rollout of computer-vision-based visual analysis. This specialized AI model scans public photos and videos uploaded to the platforms to identify physical maturity markers, analyzing broad anatomical metrics such as approximate height and bone structure to estimate an age range. Meta has explicitly clarified that this system does not utilize facial recognition protocols, meaning it does not map or identify the specific biometric identity of the individual, but rather serves as a general classification tool.
Accounts that are proactively identified as potentially belonging to minors under the platform’s minimum age requirement of 13 will face automatic deactivation. To appeal a shutdown or lift automated teen account restrictions, users must go through an official verification process involving government-issued identification or third-party digital identity estimation software. Concurrently, Meta is reinforcing its backend reporting mechanisms by augmenting human moderation teams with dedicated AI classification models to evaluate community reports regarding underage users with faster response times.
