Meta confirms smart glasses were recording without being activated

The thread treats the accidental recording issue as an AI-training-data scandal, not a one-off wearable bug. Commenters focus on the reported path from unintended footage to offshore human labeling for model development, especially because the material allegedly included private and highly sensitive scenes. Their concern is not merely that a device malfunctioned; it is that an always-available camera can turn people who never consented into raw material for a company’s AI pipeline. There is near-unanimous cynicism about Meta’s ability to earn trust for AI-enabled glasses. Several readers argue that a recording failure might be survivable for a company with a credible privacy record, but Meta’s advertising-driven data business makes every such incident look like confirmation of the worst assumptions. References to encryption promises and public trust-building do not persuade them, because commenters see those assurances as incompatible with incentives to gather and exploit more behavioral data. No one defends the product, the labeling practice, or the company’s safeguards. The practical conclusion is bleak: smart glasses will struggle for social acceptance while users and bystanders believe accidental capture can expose intimate footage to contractors and potentially feed AI training.