Meta quietly removes facial recognition code after backlash over smart glasses app
Consumer groups warn the move may be temporary and are calling for stronger biometric privacy laws and enforcement
Meta has quietly removed facial-recognition technology from software associated with its AI-powered smart glasses just days after reports revealed the company had embedded code capable of identifying strangers in public.
The apparent reversal follows a report by WIRED and analysis by privacy advocates who discovered facial-recognition systems hidden within the Meta AI companion app used with the company’s smart glasses. Researchers said the software contained components designed to convert photographs of faces into unique biometric signatures that could potentially identify individuals encountered by users in public spaces.
According to researchers at the Electronic Frontier Foundation (EFF), a June 5 update to the app appears to have removed the facial-recognition functionality. EFF’s Threat Lab said a review of the updated software found that the code responsible for triggering “Person recognized” alerts had disappeared, along with machine-learning models and databases allegedly intended to detect, digitize, and store facial biometric data.
Public scrutiny credited
Privacy advocates described the rapid removal as evidence that public scrutiny forced Meta to retreat from plans that could have dramatically expanded facial-recognition technology into everyday consumer products.
“The company’s actions speak louder than its public statements,” EFF researchers wrote in an analysis of the update. “Less than 48 hours after the public learned about the system, Meta removed nearly all traces of it from the application.”
The controversy centers on the growing capabilities of smart glasses, which combine cameras, artificial intelligence, voice assistants, and internet connectivity. Consumer advocates have long warned that adding facial recognition to wearable devices could create a network of users capable of identifying strangers without their knowledge or consent.
Meta has previously explored facial-recognition technology. The company shut down its facial-recognition system on Facebook in 2021 amid mounting legal, regulatory, and public pressure. That system had been used to identify people appearing in photos and videos uploaded to the social network.
Legal exposure over the issue
The company has faced significant legal exposure over biometric privacy issues. In 2021, Meta agreed to pay $650 million to settle a class-action lawsuit alleging its facial-recognition practices violated Illinois’ Biometric Information Privacy Act, one of the nation’s strongest biometric privacy laws.
Privacy groups say the latest episode demonstrates why voluntary corporate commitments are insufficient to protect consumers.
“While Meta appears to have backed away from this feature for now, there is no guarantee it won’t return in another form,” privacy advocates warned. Researchers noted that Meta has not publicly explained whether it plans to revive the technology in future products or what happened to any data that may have been collected during internal testing.
The incident comes amid increasing concern about the spread of facial-recognition systems across both public and private sectors. Consumer groups argue that biometric information is uniquely sensitive because, unlike passwords, faces, fingerprints, and other biological identifiers cannot easily be changed if compromised.
Advocates are renewing calls for comprehensive federal privacy legislation that would limit the collection and use of biometric data, require informed consumer consent, and give individuals the right to sue companies that misuse their personal information.
Why this matters
The controversy highlights a growing clash between rapidly advancing AI-powered consumer devices and privacy protections. As smart glasses and wearable AI products become more sophisticated, consumer advocates warn that facial recognition could transform ordinary citizens into a decentralized surveillance network unless clear legal safeguards are established.
For now, privacy groups consider Meta’s apparent retreat a victory. But they caution that the broader battle over facial recognition and biometric privacy is far from over.