Facebook could have to pay 'billions' in damages in class action lawsuit over facial recognition

• Facebook faces a class-action lawsuit over its use of facial recognition technology to tag people in photos.
• The decision by a US district judge means the company could be sued by millions of US users.
• According to the judge's order, damages could rack up to billions of dollars if Facebook loses.
• Facebook said the case has no merit.

---

Facebook could have to pay billions of dollars in damages in a class action lawsuit over facial recognition, according to various reports.

US District Judge James Donato ruled that Facebook can be sued by three Illinois users who claim the firm gathered biometric data through facial recognition without their explicit consent.They are referring to Facebook's "tag suggestions" feature, which uses facial recognition to suggest which friends to tag in a photo.

The three users have sued under an Illinois state law called the Biometric Information Privacy Act, which affords certain people protections over information such as fingerprints, retina scans, and facial recognition data.

---

Donato's decision to classify the suit as a class action means Facebook could be sued by millions of users in the state. In his order, Donato wrote: "Facebook seems to believe ... statutory damages could amount to billions of dollars."

The suit wants damages of up to $5,000 (£3,482) for every time a person's image was used without their permission. If the suit is successful, every person in the class-action could receive a payout.Facebook said the case had no merit. "We continue to believe the case has no merit and will defend ourselves vigorously," the company said in a statement, cited by Reuters.

The facial recognition feature isn't currently available in the UK. Facebook switched the feature off in Europe in 2012 after an audit by Ireland's data watchdog.

The company is now trying to roll out facial recognition technology inside the EU again, according to the Irish Times, but on an opt-in basis.