Goblin News
Goblin NewsAI news, distilled.
← Back to feed
7

Meta Faces Class Action Over Use of Facebook and Instagram Photos for AI Training and Face Recognition

SecurityTop News1 source·Sep 11

Summary

  • • Class action alleges Meta built NameTag face-recognition using FB/Instagram photos without consent
  • • Lawsuit also targets AI models Emu and Muse Image for biometric data harvesting from user photos
  • • Illinois BIPA allows up to $5K per intentional violation; national class could number in the millions
  • • NameTag was secretly embedded in Meta glasses app with 50M+ downloads, per WIRED's June 2026 report
Adjust signal

Details

Legal

Class action filed in Chicago federal court over biometric privacy violations

Parents and children from Illinois and California filed the proposed class action in federal court in Chicago, alleging Meta violated Illinois BIPA and California privacy laws by extracting biometric information from Facebook and Instagram photos without notice or consent.

Legal

NameTag accused of building faceprint database from FB/Instagram photos without consent

The suit alleges Meta's NameTag face-recognition system for smart glasses was built using Facebook and Instagram photos, converting them into biometric signatures and storing faceprints in an on-device database configured to receive updates from Meta.

Security Alert

NameTag code secretly embedded in Meta glasses app downloaded 50M+ times — per WIRED

WIRED's June 2026 investigation found NameTag code secretly embedded in Meta's glasses AI companion app, which had been downloaded more than 50 million times, even though the face-recognition feature had not been enabled for users.

Legal

AI models Emu and Muse Image also named for illegal biometric data harvesting during training

The complaint also targets Meta's generative AI systems: Emu, which Meta's CPO Chris Cox called a 'data advantage' trained on Facebook and Instagram images, and Muse Image, which drew earlier criticism for allowing users to generate images from others' public Instagram accounts before Meta removed the feature.

Stat

BIPA damages: $5K per intentional violation; national class potentially numbering in the millions

Under Illinois BIPA, plaintiffs seek $5,000 per intentional or reckless violation (or actual damages if greater) and $1,000 per negligent violation. The national class covers people whose images appeared on Facebook or Instagram, or were submitted to Meta AI systems in prompts, since September 4, 2021.

Industry Update

Meta denies wrongdoing; says NameTag has not shipped and no universal face database exists

A Meta spokesperson stated the lawsuit 'is without merit and misrepresents our work,' noting that nothing has shipped to consumers regarding NameTag and no final decision has been made on the feature. Meta reiterated it 'is not building a universal face database.'

Context

Named plaintiffs: Illinois and California families; class covers activity since Sept 4, 2021

Named plaintiffs are Francisco Alvarez and his son (Illinois residents) and Jeremy Wahl and his 10-year-old daughter (California resident). The class encompasses users nationwide whose images appeared on Facebook or Instagram or were submitted to Meta AI systems in prompts since September 4, 2021.

Source: Wired AI (trust 0.7), published 2026-09-11.

What This Means

This class action puts Meta in a legally precarious position on two simultaneous fronts: its secretive face-recognition ambitions through NameTag and its use of billions of social media photos to train generative AI models. Illinois's Biometric Information Privacy Act is one of the most powerful privacy enforcement tools in the US — per-violation damages of $5,000 for intentional violations applied across a national class potentially numbering in the millions could create enormous financial exposure. The case is made more concrete by WIRED's prior investigation revealing that NameTag code was secretly embedded in an app with over 50 million downloads, meaning users may have unknowingly been subject to biometric identification. The lawsuit adds to a growing wave of legal challenges over AI companies' use of user-generated content for training, but the biometric dimension raises the stakes considerably higher than a typical training data dispute.

Sources

Similar Events