Meta has quietly embedded face-recognition technology into its smart glasses platform through an app downloaded by millions, according to a detailed analysis of its software by WIRED. This feature, known internally as NameTag, could enable its smart glasses to identify people via biometric data stored on users' phones, raising significant privacy concerns.

The Silent Deployment of a Controversial Feature

The discovery highlights a dissonance between Meta’s public statements and its technical implementation. In April, the company claimed it would not roll out face recognition without a “very thoughtful approach.” However, NameTag code had already been integrated into the Meta AI app, which has been downloaded over 50 million times. This app is essential for using Meta’s smart glasses, including models from Ray-Ban and Oakley.

Key elements of NameTag include:

  • A face detection model
  • A face cropping model
  • A biometric encoding model

These AI components now reside on user devices, enabling the system to generate faceprints from images captured by the glasses and compare them with a database of stored faceprints on the user’s phone.

Legal and Ethical Red Flags

Meta’s renewed interest in face recognition comes amid a backdrop of legal battles and public scrutiny. In 2021, the company deleted over a billion faceprints after years of controversy surrounding its photo-tagging system. It settled a $650 million lawsuit with Illinois users and later agreed to a $1.4 billion settlement with Texas over allegations of unlawful biometric data collection.

Privacy advocates argue that embedding such technology in a mass-market wearable platform risks normalizing a capability that has previously been deemed problematic. “You're setting norms and standards by putting technology into the ecosystem,” says Joseph Jerome, a former Meta Reality Labs policy official.

A Feature on the Brink of Activation

Despite Meta’s claims that NameTag is still under consideration, independent researchers have confirmed that the feature is nearly ready for consumer use. Security researcher Cooper Quintin of the Electronic Frontier Foundation’s Threat Lab described the system as “a distributed surveillance machine” in the making.

Buchodi, an independent privacy researcher, tested the system by inputting a faceprint of the late French philosopher Michel Foucault. The app recognized it and produced a notification: “Person recognized.”

Meta has not yet enabled NameTag for users, but the infrastructure is in place. The company has also emphasized that it will not create a centralized face database, though it has not clarified whether faceprints would ever be sent back to its servers.

A Tension Between Innovation and Privacy

Meta’s internal documents suggest the company had planned to introduce NameTag in a politically sensitive environment, possibly to avoid backlash. However, the feature could address a real need for assistive technology, as blind users often rely on facial recognition to identify people in their daily lives.

Yet, the ethical and legal hurdles remain. Woodrow Hartzog, a privacy law professor at Boston University, warns that even opt-in mechanisms may not provide meaningful protection. “Framing privacy as a matter of personal choice is advantageous to businesses,” he says, “placing no meaningful limits on collection.”

As Meta continues to refine its smart glasses, the question remains: Will it prioritize user privacy, or will NameTag become the next frontier in consumer surveillance?