HISPI Project Cerebellum HISPI PROJECT CEREBELLUM

Unintended Exposure of Private Imagery from Meta AI Smart Glasses to Human Reviewers

February 27, 2026

Meta's AI-enabled smart glasses appear to have captured private content, including material that was not intended to be recorded. A Swedish investigation alleges this content, which included images and video from users' homes and daily lives, was viewed by subcontracted human reviewers in Kenya. Such exposure of sensitive information underscores the need for robust AI governance and trustworthy AI practices. For those interested in responsible AI initiatives, HISPI Project Cerebellum TAIM offers a platform to collaborate—through JOIN US—in promoting safe and secure AI development.
Alleged deployer
meta, meta-ai, sama
Alleged developer
meta, meta-ai
Alleged harmed parties
meta-ai-smart-glasses-users, people-recorded-by-meta-ai-smart-glasses, users-of-meta-ray-ban-glasses, users-of-meta-smart-glasses-in-sweden

AI governance case studies

For forensic AI governance failure analysis (TAIMScore™ case studies), browse Human Signal’s Failure Files™.

Source

Data from the AI Incident Database (AIID). Cite this incident: https://incidentdatabase.ai/cite/1418

Data source

Incident data is from the AI Incident Database (AIID).

When citing the database as a whole, please use:

McGregor, S. (2021) Preventing Repeated Real World AI Failures by Cataloging Incidents: The AI Incident Database. In Proceedings of the Thirty-Third Annual Conference on Innovative Applications of Artificial Intelligence (IAAI-21). Virtual Conference.

Pre-print on arXiv · Database snapshots & citation guide

We use weekly snapshots of the AIID for stable reference. For the official suggested citation of a specific incident, use the “Cite this incident” link on each incident page.