Meta's AI display glasses reportedly share intimate videos with human moderators (www.engadget.com)

A recent report from Sweden's Svenska Dagbladet reveals that users of Meta's AI smart glasses may inadvertently share intimate videos and sensitive financial information with human moderators outside the EU, particularly in Kenya. Employees involved in AI annotation reported witnessing nudity, bathroom activities, and credit card details while processing data captured by users. This situation raises significant concerns about privacy and compliance with EU data protection regulations, particularly given the General Data Protection Regulation (GDPR) mandates for transparency in personal data handling. For users, the issue stems from agreeing to Meta's terms of service, which permit human review of captured data to help train the company's large language models. However, the requirement to disclose sensitive information and the lack of easily accessible privacy policy details create an environment where users may not fully understand the implications of their consent. While Meta has refrained from commenting specifically on these allegations, it emphasizes that media processing adheres to its privacy policies. This incident highlights crucial discussions around ethical data handling practices and user privacy rights in the evolving landscape of AI and wearable technology.