A day later WIRED revealed that Meta had quietly embedded an unreleased facial recognition system into an app installed on more than 50 million phones. According to WIRED’s analysis of the latest version’s code, the company has removed it.
The latest version of Meta AI, a companion app to its line of sharp glasses, removes inactive software components that power the Meta system, internally called NameTag. The version released on the day of the WIRED report included several code libraries that were clearly named for facial recognition. Friday’s release includes none of them.
Andy Stone, vice president of communications at Meta, told WIRED on Monday that the feature is purely exploratory, adding, “No final decision has been made yet on what, if anything, to do here.”
On Thursday, WIRED reported that Meta had quietly integrated much of the NameTag system into the Meta AI app. Although the feature was never made publicly available, its purpose was to transform faces captured by the glasses into unique biometric signatures, commonly called facialprints, and compare them to a database of facialprints stored on the user’s device. WIRED also found that faces the system did not recognize were cropped, indexed and saved locally for future processing.
NameTag first appeared in February when New York TimesCiting internal documents, Meta said the company is developing facial recognition technology for its sharp glasses and is considering introducing the solution later this year. One memo reportedly described publishing it in a “dynamic political environment” when privacy and civil liberties advocates would be distracted. Last week, WIRED reported that much of the NameTag engine has already been built into the Meta AI app, downloaded by millions of users back in January, even though Meta has publicly stated that it has not made a final decision on facial recognition.
Following the WIRED report, Stone dismissed the findings, writing that the company could not answer questions about how the system works because “the feature does not exist.” Andrew Bosworth, Meta’s chief technology officer, called the reports “extremely misleading” and “completely unfair.”
Meta declined to answer 10 questions asked by WIRED ahead of Thursday’s publication, including: whether it has already created a database of facial profiles that NameTag uses, how long the app keeps photos and biometric data of unrecognized people stored on the user’s device, and whether this data will ever be sent back to Meta’s servers.
Additionally, Meta did not respond to a question about whether it was creating NameTags specifically for users who are blind or have low vision, nor did it respond to criticism from privacy advocates who warned that the system could allow stalkers and abusers to identify strangers in public places. It did not answer a question about whether it plans to allow users to opt-in or opt-out of the system.
The newly released version of Meta AI removes almost all traces of the feature that Meta claimed did not yet exist. The facial recognition software itself is gone, along with the code that triggers the NameTag recognition process and the “Person Recognized” alert the app would display if someone was identified. The update also removes a folder where the app may have stored cropped images and facial biometric signatures that it captured but could not identify.
Meta did not respond to WIRED’s questions about why the code was removed or whether the changes were planned before the WIRED story was published.