Meta Sued Over Alleged Use Of Facebook, Instagram Photos To Train AI
Lawsuit alleges Meta used social media photos for AI training.
Meta is facing a new lawsuit in the United States over allegations that it used users’ Facebook and Instagram photos without consent to develop facial-recognition technology for its artificial intelligence-powered smart glasses. The 66-page complaint was filed last week in the US District Court for the Northern District of Illinois by parents and their children from Illinois and California.
At the centre of the lawsuit is NameTag, an unreleased facial-recognition system reportedly developed for Meta’s Ray-Ban and Oakley smart glasses. The system is also connected to the Meta AI companion app. According to the complaint, NameTag was designed to process faces captured through the glasses, convert them into biometric signatures or “faceprints”, and compare them with facial data stored on the wearer’s phone.
The existence of NameTag was previously reported by WIRED in June after inactive code linked to the system was discovered in the Meta AI app. Analysts said the technology appeared capable of identifying people captured by the glasses by comparing facial information with other faceprints. The lawsuit alleges that Meta went beyond simply developing the technology and potentially sourced facial images from user-generated content on Facebook and Instagram.
Also Read: Delhi-NCR Bus Gang-Rape Case: NCW Seeks Delhi Police Report Within A Week
The complaint also alleges that Meta employees indicated NameTag could identify individuals through their public Instagram accounts or connections. The plaintiffs claim Meta collected facial images from its platforms, created biometric identifiers from those images and stored the information on its servers for potential use by NameTag. The lawsuit further accuses Meta of harvesting biometric information through images used to train its generative AI systems, including Emu and Muse Image.
Meta has rejected the allegations. In a statement to WIRED, a company spokesperson described the lawsuit as a “misrepresentation” of Meta’s work and said that NameTag had not been released to consumers. The company also said no final decision had been made about whether the technology would be deployed.
The proposed class action seeks to represent people in Illinois, California and across the United States who uploaded images to Meta’s platforms or provided data to its generative AI systems, dating back to September 2021. The complaint estimates that the potential national class could include millions of people, although the allegations remain subject to proceedings in court.
Also Read: Bank Safe Swept Away By Nepal Floods, 29 Detained Over Alleged Looting
