Norway wants AI glasses banned from beaches, schools, and gyms

Norway's government said on Monday it will propose a temporary ban on AI glasses in selected places, worried that people could be photographed, filmed, or audio-recorded without their knowledge. The ban could cover parks, beaches, museums, shopping centers, and public events, as well as schools, playgrounds, youth clubs, doctors' offices, swimming pools, and gyms with changing rooms. A bill goes to parliament as soon as possible. An expert group will advise on permanent national rules for body-worn technology.

"I am worried that new, powerful technology is being introduced where people risk being photographed, filmed or audio-recorded without knowing it," said Digitalisation and Public Governance Minister Torgeir Micaelsen. Note what the proposal is not: a total ban. Private use stays allowed, and so does use where others are not at risk of being filmed without consent. The government will still examine which technology the ban covers: glasses with camera and audio recording, glasses with camera and AI functions, or body-worn technology with camera, audio, and AI more broadly.

Norway is not moving in a vacuum. Dutch retailer Hans Anders has halted Ray-Ban Meta sales in the Netherlands and Belgium. Germany's conference of data protection authorities said in September that wearers are responsible for their recordings and that an indicator LED is not enough to inform the people being filmed. The Dutch authority has said filming others with camera glasses is "almost always" prohibited. In South Africa, Meta met the Information Regulator over its camera glasses; the regulator would not say the existing framework is enough. The direction of travel is consistent across all of them: the recording light does not constitute informed consent, and the wearer bears the responsibility.

For product teams, this is the consent problem we keep writing about, now wearing hardware. The pattern from Meta's Muse permission incident was that a single "allow always" setting turned into a location shared with a stranger. The pattern here is the physical version: a device that records by default in the wearer's field of view, with bystanders who never opted in. Regulators are converging on the same answer in both cases. Consent has to be explicit, per-context, and revocable, and a tiny LED is none of those things.

Three things follow if you ship anything with a camera and an agent behind it. First, design for bystander consent now, not after the bill passes. Venues will write their own policies faster than parliaments do, and schools and gyms already are. Second, build recording indicators that survive a regulator's reading: visible, unmistakable, and impossible to cover without disabling the camera. Meta's existing approach, a front LED plus software that disables the camera if the light is covered, is the current best practice, and regulators are still unimpressed. Third, watch the expert group. Temporary bans have a way of becoming permanent frameworks, and the scope question (glasses only, or all body-worn recording tech) will decide how much of the wearable roadmap this touches.

The minority government still has to pass the bill, and the expert group has not been named. But the signal is already priced in: the places where people are most vulnerable are being carved out first, and the burden of proof is shifting to the device makers. Build like the ban is coming, because in the places that matter most, it is.