Bug turns one of the biggest fears about Ray-Ban Meta smart glasses into reality
Ars Technica Scharon Harding
Some Ray-Ban Meta glasses recorded people talking without anyone saying “Hey, Meta.” Meta says a bug caused it and says it’s fixed, but the privacy fear is obvious.
Based on reporting by Ars Technica, Scharon Harding — read the original for the full story.
Summary, retelling and take written by AI under human oversight; images are AI-generated illustrations. How we work · Report an error
Some Ray-Ban Meta smart glasses were caught recording people’s conversations even when the wearer had not said the wake phrase. Meta says the problem came from a bug and says it has now been fixed. That still leaves a nasty little thought hanging over the product: if a pair of glasses can listen when nobody asked it to, trust gets expensive fast.
The reports surfaced this week, when users began posting that their glasses were recording without their consent. One Reddit user, posting as Kindly-Audience-137, said it happened across three Meta Ray-Bans. They also said the “Respond without Hey Meta” setting was turned off, which makes the behavior sound even less like user error and more like the sort of glitch that ruins a whole category of device.
That same user said the Meta AI app saved audio recordings and text transcripts of the conversations in their conversation history. That is the part that turns an abstract bug into something more concrete and more uncomfortable. It is one thing for a smart assistant to misfire. It is another for it to keep a written and audio record of private talk when people thought it was off duty.
Meta’s answer is that the bug has been fixed. Fine. But smart glasses already ask for a lot of trust, and this is exactly the kind of failure that makes people check whether the tiny computer on their face is listening when it should be minding its own business.
My take — AI-written commentary, not fact-checked reporting
This is the kind of failure that makes “always-on assistant” sound less like convenience and more like a warranty claim waiting to happen. Meta can patch bugs all it wants; people still remember which products turned private conversations into data in the first place. The industry keeps calling that innovation. A lot of users will call it creepy, which is the more honest label.
Read more about this at: Ars Technica
Related stories
Meta makes AI glasses slightly less creepy with limit on nonconsensual recording
Ars Technica · 1 month ago ·
39