A growing number of large technology companies are betting on smart glasses — eyewear that can record audio and video and run AI models on board — and the privacy questions that come with them are getting harder to wave away.
According to Wired, which frames the issue in a piece asking whether it is possible to make smart glasses that aren't creepy, the concern is scaling alongside the investment: as more big companies pour money into the category, privacy concerns are "only getting louder."
The tension is fairly simple to state. A phone camera is obvious — you hold it up, point it, and the people around you can see what you are doing. A camera built into a pair of glasses is not obvious in the same way. Add on-device AI that can listen, transcribe, or interpret what the wearer is looking at, and the device stops being a camera you occasionally use and becomes a sensor that is always present in a room with other people.
That shifts the privacy question away from the buyer. Someone who purchases a pair of AI glasses agrees to the terms of service; the stranger sitting across from them on a train does not. Wired's framing — whether a non-creepy version of this product is even achievable — suggests the design problem is not only about what data the glasses collect, but about whether bystanders can tell when they are being recorded at all.
It matters because smart glasses move the surveillance question out of our devices and into public space, where the people most affected never opted in.