Workers in Kenya Are Watching Your Most Private Moments From Meta Ray-Ban Smart Glasses

Workers in Kenya Are Watching Your Most Private Moments From Meta Ray-Ban Smart Glasses

Redacto
8 min read

Categories: Meta, News, Privacy and Security, Wearable Technology

Quick Story Summary
  • An investigation found footage from Meta Ray-Ban smart glasses can be reviewed by human contractors at Sama in Nairobi, Kenya.
  • Workers reported seeing sensitive footage captured inside homes, including people undressing and handling financial information.
  • Video can be captured when users activate the “Hey Meta” AI assistant, which sends visual data to Meta servers where it may be labeled by contractors.
  • Meta says faces are blurred, but workers claim the system often fails and people remain identifiable.
  • A 2025 update enabled the AI camera by default and removed the ability to opt out of voice recording storage.
  • A March 5, 2026 class action lawsuit and regulatory inquiries in the UK and EU are now challenging Meta’s practices.
Nairobi, Kenya Human review contractors
Apr 2025 AI camera enabled by default
Mar 5, 2026 Class action lawsuit filed

If you own a pair of Meta Ray-Ban smart glasses, there is something you should know. When you use the AI features on those glasses, the footage they capture can be sent to human workers overseas for review.

Not automated review. Human review. By real people, sitting in an office in Nairobi, Kenya.

And the footage they are seeing is not just street corners and restaurant menus.

What Did the Investigation Find?

A joint investigation by Swedish newspapers Svenska Dagbladet and Göteborgs-Posten, published on February 27, 2026, revealed that footage from Meta’s smart glasses is being reviewed by workers at a company called Sama. Sama is a data annotation subcontractor hired by Meta.

The journalists traveled to Kenya and interviewed multiple Sama employees. These workers are called data annotators. Their job is to watch video clips captured by the glasses and label objects and actions in the frame, helping Meta’s AI learn to interpret the real world.

What they described seeing was shocking.

Workers told reporters they had reviewed footage of people using the bathroom, getting undressed, engaged in sexual activity, and handling bank cards – all captured inside people’s own homes. One worker described seeing a man place his glasses on a bedside table and leave the room, only for his partner to walk in and change clothes – apparently unaware the glasses were still transmitting.

Meta says faces in footage are automatically blurred before reaching contractors. But multiple workers said the blurring system frequently fails, leaving faces clearly visible.

Workers also said they felt unable to raise concerns. As Futurism reported, employees described a culture where questioning the work could cost them their jobs.

How Does Your Footage Get Sent to Kenya?

The glasses capture footage in two ways.

First, you can manually record photos or video by tapping a button on the frame. That content stays on your phone unless you actively share it.

Second- and this is where the problems begin – the glasses have a built-in AI assistant activated by saying “Hey Meta.” When you ask the AI to identify an object, describe a scene, or translate text, the glasses capture visual data and send it to Meta’s servers. From there, it can be routed to contractors like Sama for labelling and review.

Most users have no idea this is happening.

Meta Quietly Made Things Worse in 2025

In April 2025, Meta updated its Smart Glasses policy to make the “Meta AI with camera” feature active by default. Unless you manually turn off “Hey Meta,” the AI camera is always enabled.

The same update removed the ability to opt out of voice recording storage. Your voice recordings are now stored on Meta’s servers for up to a year by default. You can delete them manually through the companion app, but you cannot stop them from being collected in the first place.

The glasses do flash a small LED when recording. But the light is easy to miss, and it gives no indication of whether your footage is being sent to a third party for human review. Additional, as we’ve covered in the past, there are numerous aftermarket modifications to hide the LED light.

Meta Is Now Facing a Class Action Lawsuit

On March 5, 2026, a class action lawsuit was filed in federal court in San Francisco against Meta and Luxottica of America, a subsidiary of Ray-Ban parent company EssilorLuxottica.

The lawsuit was brought by the Clarkson Law Firm on behalf of two plaintiffs: Gina Bartone of New Jersey and Mateo Canu of California.

Their argument is straightforward. Meta marketed the glasses with slogans like “designed for privacy, controlled by you.” The lawsuit alleges that no reasonable person would understand those promises to mean their private footage from inside their home would be watched and catalogued by overseas workers.

The complaint seeks monetary damages, punitive penalties, and a court order requiring Meta to change its practices. With over seven million pairs sold in 2025, the potential class is massive.

Meta has not commented on the lawsuit. A spokesperson told TechCrunch that media stays on the user’s device unless they choose to share it, and that when content is shared with Meta AI, contractors may review it to improve the experience. The company pointed to its terms of service and privacy policy.

UK and EU Regulators Are Investigating

The fallout has not been limited to the United States.

The UK’s Information Commissioner’s Office (ICO) has formally written to Meta demanding information about how the company meets its data protection obligations. The ICO stated that devices processing personal data should give users control and provide clear transparency about how their information is used.

At the EU level, multiple Members of the European Parliament have submitted questions to the European Commission about whether Meta’s data transfer practices comply with GDPR. The Irish Data Protection Commission, which oversees Meta in the EU, may also open its own inquiry.

What This Means for Your Data on Meta’s Platforms

The Ray-Ban story is alarming on its own. But it also illustrates a broader reality about how Meta handles data.

Every post, photo, comment, and message you leave on Instagram or Facebook is part of Meta’s data ecosystem. Meta’s own AI terms of service reserve the right to use your interactions to improve its AI products. The same AI systems. The same data pipeline.

You cannot undo footage that has already been sent from a pair of smart glasses. But you can take control of the content you have voluntarily posted over the years.

How Redact Can Help You Take Back Control

Redact is a free desktop and mobile app that lets you bulk delete your posts, comments, likes, and messages from Instagram, Facebook, Twitter/X, Reddit, Discord, and over 30 other platforms, with filters for date ranges, keywords, engagement and more.

Given what this investigation has revealed about how Meta uses personal data, reducing your digital footprint on Meta’s platforms is a practical step anyone can take right now.