Meta Sued Over Training Data for Its AI and Face-Recognition Systems
Meta is facing a class-action lawsuit in federal court, alleging illegal use of Facebook and Instagram photos to train its NameTag face-recognition system and generative AI models like Emu and Muse Image without user consent.
Intelligence analysis by Gemini 2.5 Flash
A new lawsuit accuses Meta of violating privacy laws in Illinois and California by extracting biometric information from user photos on its social media platforms. The plaintiffs claim this data was used to develop AI technologies, including an unreleased face-recognition system for smart glasses and image-generation tools, without proper notice or consent from the individuals involved.
Imagine you take pictures with your family and share them on a social media app. Now, imagine that app secretly uses those pictures to teach a computer how to recognize everyone's faces, even for new smart glasses, without ever asking you if that's okay. A group of parents and kids are suing the company, saying it's like the company took their special face information without permission, which is against the rules, and they want the company to stop and pay for what they did.
Analysis
The recent class-action lawsuit against Meta centers on the alleged unauthorized use of user photos from Facebook and Instagram to fuel its burgeoning AI initiatives. Filed in federal court in Chicago by parents and their children from Illinois and California, the complaint asserts that Meta illegally extracted biometric information from these images without obtaining explicit consent or providing adequate notice. This legal challenge underscores the growing scrutiny over data collection practices by tech giants, especially as they integrate advanced AI capabilities into their products and services.
NameTag
The lawsuit specifically targets NameTag, an unreleased face-recognition system designed for Meta's smart glasses. WIRED previously reported that code for NameTag was embedded in the Meta glasses AI companion app, which had been downloaded over 50 million times. Although not enabled for users, the system was reportedly designed to convert faces captured by the glasses into biometric signatures and compare them against a database of "faceprints" stored on the user's phone, with updates potentially coming from Meta. The plaintiffs allege these faceprints are derived from Facebook and Instagram images, citing internal claims by Meta employees that NameTag could recognize individuals through their Meta connections or public Instagram accounts, and a company patent describing face matching against profile photos.
Meta has publicly stated it is "not building a central face database" and that "nothing has shipped to consumers" regarding NameTag, with no final decisions made. However, the complaint highlights Meta's past reluctance to clarify whether NameTag would be opt-in or how faceprints would be retained. The company's CTO, Andrew Bosworth, initially dismissed WIRED's reporting as "incredibly misleading" but later described NameTag on a podcast as a potentially "great feature" for recognizing people a wearer had previously met and asked the device to remember. This apparent contradiction fuels the plaintiffs' claims that Meta has been less than transparent about its intentions and capabilities concerning biometric data.
Emu and Muse Image
The lawsuit also extends to Meta's generative AI systems, Emu and Muse Image. Meta's chief product officer, Chris Cox, has openly referred to Facebook and Instagram platforms as a "data advantage" for training its AI systems, acknowledging the use of large quantities of images and text from these platforms for Emu. The complaint alleges that this training process illegally harvested biometric information from individuals featured in the images. Muse Image, which was released briefly, faced significant criticism for allowing users to generate images based on other people's public Instagram accounts, a feature Meta quickly removed, admitting it had "missed the mark." These incidents collectively paint a picture of Meta leveraging user data for AI development in ways that allegedly bypass privacy safeguards and user consent.
Illinois Biometric Information Privacy Act
The legal action is primarily grounded in Illinois' Biometric Information Privacy Act (BIPA) and California privacy laws. BIPA is one of the strictest biometric privacy laws in the United States, requiring companies to obtain informed consent before collecting, storing, or using biometric data. The plaintiffs are seeking substantial damages under BIPA, including $5,000 for each intentional or reckless violation and $1,000 for each negligent violation, in addition to injunctive relief. The complaint estimates the national class could number in the millions, suggesting potentially massive financial implications for Meta. This is not Meta's first encounter with BIPA; in 2020, the company settled an Illinois class action for $650 million over an earlier face-recognition system, which it subsequently shut down. In 2024, Meta also agreed to pay Texas $1.4 billion to resolve similar allegations of unlawfully collected biometric data. These prior settlements indicate a pattern of legal challenges for Meta regarding its handling of biometric information, reinforcing the plaintiffs' argument that the current case is part of a longer history of privacy violations.
Key points
- Meta is being sued in a class-action lawsuit for allegedly using Facebook and Instagram photos to train its AI systems without consent.
- The lawsuit targets NameTag, an unreleased face-recognition system for smart glasses, and generative AI models like Emu and Muse Image.
- Plaintiffs allege Meta violated Illinois and California privacy laws by extracting biometric information from user images.
- Meta has previously faced and settled large lawsuits over its handling of biometric data, including a $650 million settlement in Illinois and a $1.4 billion settlement in Texas.
- The complaint estimates the proposed national class could include millions of people, seeking substantial damages for each violation.
Meta faces significant financial penalties and reputational damage if found liable, potentially leading to billions in fines and stricter regulations on how it can use user data for AI training. This could force a complete overhaul of its data collection practices and slow down its AI development efforts.
Market signals
- META The proposed class action seeks significant damages and injunctive relief, potentially impacting Meta's financial standing and operations, as evidenced by past large settlements.
AI-generated analysis of potential market relevance. Not financial advice.
