mart glasses used to be a clunky sci-fi gimmick, but now they sit quietly on your face as stylish eyewear while potentially capturing your private conversations without your consent. Some Ray-Ban Meta smart glasses were recording people's conversations without the wearer knowing, as highlighted in reports by Android Authority on Tuesday. The recordings occurred without users uttering the required activation phrase ("Hey, Meta"), due to a bug that Meta claims has now been patched. A Reddit user posting under the handle Kindly-Audience-137 reported that this happened across three separate pairs of Meta Ray-Bans, despite having the "Respond without Hey Meta" setting explicitly turned off.
For anyone considering strapping a camera and a microphone to their face, this glitch turns every theoretical privacy nightmare into a tangible reality. The affected user noted that the Meta AI app saves audio recordings and text transcripts of these conversations directly in the user's conversation history. This means private moments were actively transcribed and stored in the cloud without intent. As Kindly-Audience-137 observed, customers should not have to wonder whether their private conversations are being harvested when they never even touched their glasses' AI assistant.
This issue has been resolved. People can manage, review, and delete voice logs of their interactions with Meta AI via settings at any time.
Unanswered Questions and the Cloud Trap
As Meta stated in an official comment to reporters, the issue has been resolved. However, the company has offered zero transparency regarding what actually caused the Ray-Ban Metas to record people without consent. Reddit user Kindly-Audience-137 stated they first noticed this behavior immediately following a recent firmware update, but Meta has steadfastly refused to confirm whether a faulty patch triggered the bug. We still have no idea how many devices were compromised, and Android Authority noted it was unable to replicate the problem before Meta silently pushed out the fix.
Long before this bug came to light, privacy advocates had already raised red flags over Meta’s smart glasses and competing hardware. Beyond the baseline unease of walking around with an always-on camera and microphone, Meta’s frames blurred the line completely by looking entirely normal. That inconspicuous design is precisely what makes unauthorized recording terrifying in practice—you never know if the person across the table is streaming your dinner conversation straight to a cloud database.
In our view, a quiet software patch does not erase the fundamental vulnerability. If a background glitch can bypass wake words and ship private dialogue to a server without warning, the entire premise of face-worn surveillance hardware needs a hard reality check. Until manufacturers offer local-only storage and verifiable hardware kill switches, wearing these gadgets means trusting a tech giant's code not to malfunction while pointing directly at your face.