Meta has quietly removed face recognition software from its smart glasses companion app just one day after WIRED revealed the company had embedded an unreleased facial identification system into an app installed on more than 50 million phones. The removal raises fresh questions about the company’s privacy practices and its willingness to deploy surveillance technology without user consent.
The latest version of Meta AI, which works with the company’s line of smart glasses, strips out the unactivated software components that powered what Meta internally called “NameTag.” The previous version included several code libraries explicitly named for face recognition. Friday’s release contains none of them.
The swift removal highlights the growing tension between tech companies’ desire to push the boundaries of AI-powered surveillance and mounting privacy concerns from regulators and civil liberties groups. Meta’s decision to initially include the dormant code, then quickly remove it after public scrutiny, suggests the company may be testing how far it can go with facial recognition technology before facing backlash.
Andy Stone, Meta’s vice president of communications, told WIRED that the feature remains purely exploratory. “No final decision has been made on what to do here, if anything,” Stone said Monday. However, the company declined to explain why the code was removed or whether the changes were planned before WIRED’s investigation.
The NameTag system was designed to convert faces captured by Meta’s smart glasses into unique biometric signatures, then compare them against a database stored on the user’s device. WIRED’s analysis also found that faces the system couldn’t recognize were cropped, indexed, and stored locally for future processing.
This discovery is particularly significant given Meta’s troubled history with facial recognition. The company previously faced a $650 million settlement over its use of facial recognition technology on Facebook without user consent. Privacy advocates have warned that putting similar capabilities into smart glasses could enable stalking and harassment by allowing users to identify strangers in public.
Meta executives initially dismissed WIRED’s findings. After the original report, Stone claimed the company couldn’t answer questions about how the system would work because “the feature does not exist.” Andrew Bosworth, Meta’s chief technology officer, called the reporting “incredibly misleading” and “absolutely dishonest.”
Yet the code removal tells a different story. The newly released version of Meta AI removes nearly all traces of the facial recognition system, including:
- The face-recognition software itself
- Code that ran the NameTag recognition process
- A “Person recognized” alert the app would have displayed
- A folder where the app stored cropped images and biometric signatures of unidentified faces
Some fragments of the NameTag system remain in the latest version, including an internal debug menu label and a dormant link meant to open a recognized person’s profile. These leftover pieces point to parts of the system that are no longer there.
The incident underscores the challenges facing lawmakers trying to regulate AI and privacy in an era of rapid technological advancement. Kade Crockford, director of the technology for liberty program at the American Civil Liberties Union of Massachusetts, says the removal doesn’t undo Meta’s original decision to ship the code.
“Meta’s sneaky tactics in slipping the face-recognition code into its smart glasses show exactly why data privacy bills need the teeth of strong enforcement,” Crockford says. “Companies like Meta prioritize their bottom line, so lawmakers need to speak in the only language its C-suite understands.”
The timing of NameTag’s development is also noteworthy. The system first surfaced in February when The New York Times reported that Meta was considering launching face recognition as soon as this year. One internal memo reportedly described releasing it during a “dynamic political environment,” when privacy advocates would be distracted by other issues.
Meta’s smart glasses, developed in partnership with Ray-Ban, have already raised privacy concerns due to their ability to record video and take photos discretely. Adding facial recognition capabilities would significantly amplify these concerns, potentially turning the glasses into a tool for mass surveillance.
The company has not responded to key questions about the NameTag system, including whether it had already created a database of face profiles, how long the app retains biometric data, and whether that information would ever be sent to Meta’s servers. The company also hasn’t said whether it was building NameTag specifically for blind or low-vision users, or whether users would be able to opt out.
This episode reflects broader industry trends around AI development and deployment. Tech companies increasingly build and test AI capabilities internally, sometimes embedding them in consumer products before announcing them publicly. This approach allows for rapid development but raises questions about user consent and transparency.
As AI becomes more powerful and pervasive, the NameTag controversy may serve as a preview of future battles over surveillance technology. Privacy advocates argue that stronger legal protections are needed to prevent companies from deploying invasive AI systems without explicit user consent.