Workers in Nairobi stared at screens filled with strangers’ most private moments. Footage from Ray-Ban Meta smart glasses showed people undressing. Using the toilet. Having sex. And Meta just cut their jobs.
The trouble started in February, when Swedish newspapers Svenska Dagbladet and Göteborgs-Posten published a joint investigation. Reporters interviewed over 30 employees at Sama, Meta’s Kenyan subcontractor. Nine of those workers handled video from the glasses. One described a clip where a man left his Ray-Ban Meta pair on a bedside table. His wife walked in. Changed her clothes. “I don’t think they know, because if they knew they wouldn’t be recording,” the annotator said.
Another worker saw sex scenes. “Someone is wearing them, having sex. That is why this is so extremely sensitive.” Bank cards flashed in frames. People watched porn with glasses on. Chats revealed crimes, protests. “We see everything—from living rooms to naked bodies,” a Sama employee told the Swedes. Algorithms blurred faces first. But they failed in bad light. Bodies stayed visible.
Sama’s data annotators label clips to train Meta’s AI. They draw boxes around objects—pots, signs, cars. Transcribe speech. Verify answers. Videos hit Meta servers in Sweden, Denmark, Ireland. Some get manual review. Users consent in fine print, Meta says. The glasses, a hit with seven million sold, light up when recording. But forget them on a table? Problem.
Less than two months later, Meta paused work with Sama. Then ended the contract. 1,108 Kenyan jobs gone. Sama claims it met all standards. “At no point were we notified of any failure,” the company said in a statement to BBC News. Workers disagree. Naftali Wambalo of Africa Tech Workers Movement calls it retaliation. “What I think are the standards they are talking about here are standards of secrecy.”
Meta pushes back. “We have decided to end our work with Sama because they don’t meet our standards,” a spokesperson told BBC. Photos and videos stay private unless shared with Meta AI. Humans review shared content—with consent—to boost performance. Filters protect privacy first.
But the damage spread. UK’s Information Commissioner’s Office called the Swedish report “concerning.” They wrote Meta, demanding compliance details under data law, per another BBC article. Kenya’s Data Protection Commissioner launched a probe. Class-action suits followed. One alleges false privacy promises like “designed for privacy, controlled by you.” Plaintiffs claim they didn’t expect overseas eyes on intimate clips.
And the firings? Sama workers believe speaking out sealed their fate. Mercy Mutemi, a lawyer with Oversight Lab, sees a shaky AI foundation. “We’ve been told that this is our entry route into the AI ecosystem. This is a very flimsy foundation.” Naipanoi Lepapa, the Kenyan journalist who aided the Swedish probe, highlighted the human cost.
Meta’s glasses keep selling. Voice commands answer questions about the world. Hands-free. Cool. Yet users leave them lying around bedrooms. Bathrooms. The AI needs real-world data to learn. Humans provide it—at a price. Contractors feel the trauma. “You understand that it is someone’s private life you are looking at, but at the same time you are just expected to carry out the work,” one told investigators. Question it? You’re gone.
Regulators circle. Sales boom anyway. Privacy policies bury the truth in legalese. Consent? Users click through without reading. Footage flows to Nairobi. Labels get applied. AI improves. But trust erodes. One blurred face at a time.
So where does this leave wearables? Ray-Ban Meta pairs promise freedom. They deliver surveillance too. Contractors paid pennies to watch. Fired when they talk. Meta moves on to new partners. The cycle repeats.
WebProNews is an iEntry Publication