The promise of the Ray-Ban Meta smart glasses was straightforward: a seamless integration of artificial intelligence into our daily lives, making technology feel natural, almost invisible. However, a new wave of allegations from workers in Kenya is shaking the foundations of trust, revealing a grim reality behind the training of the company's AI models. The reports, surfacing through investigations into labor conditions in the content moderation sector, speak of access to highly personal, even nude moments of users, recorded without their full awareness.
The "Human" Machine Behind Artificial Intelligence
For the AI in Ray-Ban glasses to function, it requires vast amounts of data so the system can "learn" to recognize objects, faces, and contexts. What Meta often fails to emphasize in its marketing campaigns is that this process is not fully automated. Thousands of workers in the "Global South," primarily in countries like Kenya, India, and the Philippines, work as data labelers. Their job is to watch video clips and photos captured by the glasses and categorize them.
According to the allegations, these workers encountered material that should never have left the private sphere of the users. From private moments in bathrooms to intimate encounters, the glasses' cameras captured content that users believed was protected. The problem is exacerbated by the fact that many users do not realize that selecting "product improvement" in the settings essentially grants human reviewers permission to view their footage.
The Ethical Gap and Exploitation
This case highlights two major issues: the violation of privacy and labor exploitation. Data labelers in Kenya, often paid meager wages, are forced to watch traumatic or inappropriate content without adequate psychological support. Meta, through subcontractors like Sama, appears to have created a digital "assembly line" where the privacy of wealthy Western users is sacrificed for development, while workers in the developing world bear the moral and psychological burden.
"We don't just see data. We see lives. We see people in their most vulnerable moments, unaware that someone thousands of miles away is watching them," says one of the whistleblowers.
Meta responds that the process is necessary for the safety and accuracy of AI and that data is anonymized. However, anonymization in a video showing the interior of a home or a person's face is practically impossible, as visual elements betray the individual's identity.
Legal Implications and the Future of Wearables
This revelation is expected to trigger intervention from regulatory authorities, particularly in the European Union, where the General Data Protection Regulation (GDPR) sets strict limits on the collection of biometric and personal data. If it is proven that Meta did not adequately inform users that humans—and not just algorithms—would have access to their material, fines could reach billions of euros.
The bigger question remains: can we trust a device we wear on our faces that has a camera and a microphone? Wearable technology promises to free us from phone screens, but if the price is turning our private lives into training material for multinationals, then society may not be ready to pay it. The need for transparency is no longer a theoretical requirement but an urgent necessity for the survival of digital ethics.
Conclusions and Challenges
Meta is at a crossroads. On one hand, competition with Apple and Google pushes it to release new AI features rapidly. On the other, the disregard for privacy could lead to a massive consumer backlash. The case in Kenya is a reminder that artificial intelligence is not a "magical" entity born in a vacuum, but a product built on human labor and, often, the violation of the most fundamental human rights.
