A class action filed against Meta accuses the company of using photos posted on Facebook and Instagram to extract biometric information and develop facial recognition and generative artificial intelligence systems without the consent required by law. The lawsuit was filed on September 4 in a federal court in Illinois, in the United States.
The lawsuit, brought by residents of Illinois and California, claims that images of users and other people depicted on the platforms were used to train and test NameTag, a facial recognition system developed for Meta's smart glasses, as well as image generation models such as Emu and Muse Image. The allegations have not yet been ruled on.
According to the complaint, the systems would turn facial features into numerical representations capable of identifying a person. The plaintiffs allege that this process creates and stores information considered biometric and that Meta did not provide adequate notice or obtain consent for this purpose.
Meta itself acknowledges that public content from its services is involved in AI development. The company has already stated that public Facebook and Instagram posts, including photos and texts, were used in the training of generative models. The legal dispute, however, goes beyond the use of the images and questions whether their processing created and stored biometric identifiers protected by law.
Lawsuit could cover millions of people in the U.S.
The plaintiffs seek to represent a nationwide class composed of people in the United States whose images were uploaded to Facebook, Instagram or Meta's generative models, as well as specific groups in Illinois and California. The period indicated in the lawsuit begins on September 4, 2021, and the complaint estimates that the nationwide class could reach 1 million people.
In Illinois, the lawsuit invokes the Biometric Information Privacy Act, known as BIPA. The plaintiffs are seeking, among other measures, damages of up to US$ 5,000 per violation deemed intentional or reckless and up to US$ 1,000 per negligent violation, as well as measures to restrict practices considered illegal.
Part of the case also involves NameTag, technology found in June in the app used with Meta's smart glasses. The feature was not enabled for consumers, but the code analyzed allowed faces captured by the cameras to be converted into biometric signatures and compared with records stored on the device. Meta removed the code from the app after the system was disclosed.
The company rejected the allegations. In a statement to Wired, it said the lawsuit “has no merit” and distorts its work, said it is transparent about the use of information to develop its AI products and stressed that NameTag was not released to consumers. The company also said it is not building a universal face database.
The dispute comes after other cases involving biometrics. In 2021, Meta announced the shutdown of Facebook's facial recognition system and the deletion of templates from more than 1 billion people. In 2024, it agreed to pay US$ 1.4 billion to Texas to settle allegations related to the unauthorized collection of biometric data.
