Meta's AI Glasses Reportedly Expose Private Lives to Distant Reviewers

An investigation by two Swedish newspapers alleges that Meta’s popular AI-powered smart glasses are sending deeply personal footage to human contractors in Nairobi, Kenya for review. According to Svenska Dagbladet and Göteborgs-Posten, these reviewers have seen videos recorded by the glasses that include people in bathrooms, engaged in intimate acts, and in various states of undress.

The report has already sparked a proposed class action lawsuit in the United States, which accuses Meta of false advertising. The suit points to the company’s privacy claims for the glasses, arguing Meta concealed that using AI features could mean "a stranger halfway around the world watching the most private moments of a person’s life."

Workers in Kenya, employed as AI annotators to label data for training, described the content they handle. "We see everything — from living rooms to naked bodies," one told reporters. While a former Meta employee stated faces are automatically blurred, contractors said the system frequently fails, leaving identities and sometimes even bank card details exposed.

Meta’s collaboration with EssilorLuxottica has been a commercial success, with over 7 million units sold in 2025 alone. However, privacy changes last year kept the camera-enabled 'Hey Meta' assistant active by default and required voice recordings to be stored in the cloud.

In a statement, Meta spokesperson Tracy Clayton said media from the glasses "stays on the user’s device" unless shared. She confirmed contractors sometimes review shared data to improve AI, adding that steps are taken to filter identifying information. The UK’s Information Commissioner’s Office is now questioning Meta about these practices, as privacy advocates warn of broader risks if facial recognition is added to the devices.