EFF and Wired report that Meta has shipped facial-recognition code in the software for its always-on smart glasses, potentially affecting people both using the glasses and those seen by them. EFF says static analysis confirmed code that stores faceprints as 2,048-value templates and compares newly seen faces against a local database; researchers also showed the feature could be triggered in testing by manually adding a face in debug mode, though it is not yet exposed as a consumer setting.
Why it matters: This is a significant surveillance and privacy story because it suggests consumer wearables may already contain hidden person-identification features before any public rollout. People considering Meta glasses should weigh the privacy risk, and policymakers and civil-society groups may press Meta for transparency, safeguards, or limits before deployment.
Rindala Alajaji
2026.06.08
97% relevant
This is a direct update to the same underlying event: after the earlier reporting that Meta's smart-glasses app contained active facial-recognition code, EFF now says Meta's June 5 app update removed the face-recognition components, including recognition alerts, biometric-signature handling, and related models/databases.
Cooper Quintin
2026.06.04
100% relevant
This article establishes a new story by documenting previously unreported facial-recognition functionality in Meta's smart-glasses software, with independent technical confirmation rather than merely opinion or advocacy.
← Back to all stories