Meta Ray-Bans trigger fresh backlash over human review

Consumer AI hardware is running into the same trust wall as social platforms: weak transparency around who sees user data and when.

• –Human-in-the-loop labeling for wearable AI creates a bigger consent problem than typical chatbot logs because bystanders are captured too.
• –Meta’s policy language appears broad enough to permit manual review, but user understanding of that tradeoff is likely low.
• –This kind of reporting increases regulatory pressure in privacy-heavy jurisdictions and can slow smart-glasses adoption.
• –For AI product teams, the lesson is clear: make capture, retention, and human-review controls explicit and default-safe.