Meta quietly embedded an unreleased facial recognition system called NameTag into its Meta AI app for smart glasses, which was installed on more than 50 million phones as of early 2026 [1, 2, 3]. The system converted faces captured by the glasses into biometric faceprints and compared them to a faceprint database stored locally on users’ devices [1, 2]. Faces not recognized by the system were cropped, indexed, and saved locally for potential future processing [1, 2].
Development of NameTag reportedly began in early 2026, with the code integrated into the Meta AI app as early as January 2026 [1, 2, 3]. However, the facial recognition feature was never publicly enabled or released to users [1, 2]. Meta had previously stated it had made no final decision on whether to activate face recognition for its smart glasses [1, 2].
The presence of NameTag was revealed in a Wired report published on June 4, 2026 [1, 2, 3]. One day later, Meta released an updated version of the Meta AI app that removed the NameTag code following the report [1, 2]. Meta’s VP of communications Andy Stone described the feature as purely exploratory and said, “No final decision has been made on what to do here, if anything.” [1]
Meta’s CTO Andrew Bosworth called the reporting on NameTag “incredibly misleading” and “absolutely dishonest” [1]. Meta declined to respond to multiple questions about the faceprint database status, data retention policies for unrecognized faces, and whether any data was transmitted to Meta’s servers [1, 2].
Privacy advocates criticized Meta over the NameTag system, raising concerns about surveillance and data privacy [3]. The Meta AI app initially embedding the facial recognition code on millions of devices without public disclosure intensified these worries.
Meta’s update removing the facial recognition system marks the most recent concrete step. The company has not announced any further plans regarding NameTag or facial recognition features for its smart glasses to date [1, 2].
