AI Age Estimation: How Meta Uses Body Size and Bone Structure to Verify User Age (2026)

AI Age Estimation: How Meta Uses Body Size and Bone Structure to Verify User Age (2026)

Meta is deploying AI to estimate user age on Instagram and Facebook by analyzing body height, posture, and skeletal structure—without using facial recognition. The system, confirmed by internal sources and reported by The Decoder, targets users who may have falsified their age during sign-up, aiming to comply with child safety laws like COPPA and the DSA. This marks a pivotal shift toward non-facial biometrics in social media moderation.

How Body Size AI Works: Beyond Facial Features

Meta’s AI scans uploaded photos and videos to detect developmental markers linked to adolescence or adulthood. Unlike facial recognition, it focuses on metrics like shoulder width, limb proportions, and spinal alignment visible in full-body or partial shots. These cues are trained on anonymized datasets of known-age individuals, enabling the model to infer biological age even when faces are obscured.

Regulatory Challenges in the EU and US

While COPPA and the EU’s Digital Services Act require age verification, they don’t explicitly permit non-consensual biometric analysis. Privacy advocates warn that body size and bone structure are sensitive biometric data under GDPR and proposed U.S. bills like the DELETE Act. The FTC has signaled interest in reviewing Meta’s approach, calling it a "gray zone" in biometric regulation.

False Positives and Accuracy Concerns

Internal testing revealed the system misidentifies up to 18% of adults as minors—particularly tall adolescents or muscular adults. Conversely, some underage users with early development evade detection. Meta has not published validation metrics, raising concerns about fairness and due process. Users currently have no way to appeal or correct inaccurate age flags.

Alternatives to Facial Recognition: A Growing Ecosystem

Tools like Vidnoz AI and PiktID demonstrate how easily AI infers identity from body metrics—even in edited or filtered images. FasterCapital’s research on AR filters shows platforms increasingly embed real-time biometric analysis, often without disclosure. These technologies, marketed as fun or marketing tools, normalize surveillance-grade inference.

The Surveillance vs. Safety Dilemma

Meta claims its goal is child safety, reducing exposure to inappropriate content and targeted ads. But experts argue that continuous, non-consensual biometric scanning sets a dangerous precedent. As AI-driven identity inference becomes standard, the line between protection and profiling blurs—especially when users can’t opt out or understand how their data is used.

Without transparent audits, user consent, or regulatory clarity, Meta’s AI age estimation may solve one problem while creating broader ethical and legal risks. As platforms race to comply with 2026’s evolving child safety laws, the question isn’t just whether the tech works—but whether society will allow it.