The Wrong Pocket, the Wrong Moment
A guy walks into a café in Oslo, sits down, and takes off his glasses to clean them. Across the room, a woman notices the small LED light on the frames had been glowing a second ago. She doesn’t know if she was just recorded. She doesn’t know who has that clip, or where it’s going. She has no way to find out.
That is the reality Meta’s Ray-Ban smart glasses — and whatever Snap, Apple, and a dozen smaller players are pushing into the same space — have quietly built into the world. And Norway’s digital minister Karianne Tung just said, out loud, what a lot of people have been thinking: maybe we should ban the damn things from public spaces until we figure out what they’re actually doing to privacy.
The country is considering legislation to regulate or ban camera-enabled wearable headsets, specifically naming Meta and Snap devices. Tung’s quote is worth sitting with: “We must prevent this equipment from being used to monitor other people in public places.”
This isn’t a fringe rant. It’s a government minister responding to a technology that has, in roughly two years, gone from niche gadget to something people quietly wear everywhere — and that has already generated a lawsuit, a UK regulator investigation, a class-action complaint, and the unfortunate nickname that stuck: pervert glasses.
What These Glasses Actually Are
Let’s be clear about what we’re talking about. The Ray-Ban Meta glasses look like ordinary eyewear. That’s the entire product strategy. They have cameras built into the frames, microphones that capture audio, and the ability to livestream what you’re seeing directly to Facebook or Instagram. Meta’s latest generation added a display — a tiny screen in the corner of one lens — which makes the glasses even more capable and even harder to immediately identify as a recording device.
The capture indicator is a white LED on the front that blinks when recording. Meta says it can’t be disabled and that covering it disables the camera. That’s something, technically. But a blinking light the size of a pinhead on a moving person’s face is not the same thing as the unmistakable act of someone holding up a phone.
Gene Kang, a privacy lawyer quoted in Fortune’s coverage of the backlash, put it precisely: “If you’re holding up your phone to somebody’s face, they’re going to know.” With glasses, there’s no such implied consent. The person being recorded often has no idea it’s happening.
That’s the core design tension. Smart glasses are built to be invisible as surveillance devices. The people around the wearer are not the customers. They’re the raw material.
The Lawsuits and Investigations Already Happening
This isn’t hypothetical harm. In March 2026, Meta was hit with a class-action lawsuit — Bartone v. Meta, filed by the Clarkson Law Firm — alleging the company violated privacy laws and engaged in false advertising. The plaintiffs, Gina Bartone of New Jersey and Mateo Canu of California, pointed to marketing that promised the glasses were “designed for privacy, controlled by you” and “built for your privacy.”
Then Swedish newspaper SVD investigated and found that workers at a Kenya-based subcontractor were reviewing footage from customers’ glasses — including sensitive content like people undressing, having sex, and using the toilet. Meta said it blurred faces, but multiple sources told the BBC the blurring didn’t consistently work. The UK’s Information Commissioner’s Office opened an investigation.
Meta’s response, via spokesperson Christopher Sgro: “Ray-Ban Meta glasses help you use AI, hands-free, to answer questions about the world around you. Unless users choose to share media they’ve captured with Meta or others, that media stays on the user’s device.”
The phrase “when people share content with Meta AI, we sometimes use contractors to review this data” is in the privacy policy. But the lawsuit makes a reasonable point: if you market something as privacy-preserving and the fine print says overseas contractors may review your intimate footage, that’s a gap between promise and practice that consumer protection law should care about.
Then there’s the facial recognition question. In February 2026, the New York Times reported that Meta planned to add facial recognition to its smart glasses. WIRED later found that code for a feature internally called “NameTag” was already embedded in Meta’s AI app — software installed on millions of phones — capable of identifying faces captured by the glasses’ camera, sending recognized faces a notification, and saving unrecognized ones to a “pending” folder. More than 70 advocacy groups, including the ACLU and the Electronic Privacy Information Center, demanded Meta scrap the feature. Meta publicly said it wouldn’t roll out facial recognition without a “very thoughtful approach.”
Norway’s Tung explicitly flagged this as part of the regulatory concern: the combination of AI, cameras, microphones, and potential facial recognition built into objects people wear all day.
The Arms Race Nobody Asked For
Here’s where things get darkly interesting. People have noticed. And they’re fighting back in ways that say a lot about where we are as a society.
A professor named Dr. Yves Jeanrenaud built an open-source app called Nearby Glasses that tries to alert you when Meta Glasses are nearby. His GitHub readme calls the glasses “an intolerable intrusion, consent neglecting, horrible piece of tech that is already used for making various and tons of equally truly disgusting ‘content’.” He also notes, responsibly, that it’s an imperfect solution and won’t catch everything.
Fortune reported people taking extreme measures to avoid being filmed — physical blocking, avoiding certain areas, modifying their behavior in public. The article quoted Harvard computer science professor Jim Waldo: “We are living in weird times… The technology is changing. It’s the combination of the Meta Glasses with facial recognition, AI, and a number of other sorts of technologies that are all coming together and putting us in an environment that we just aren’t prepared to deal with yet.”
Read that again: we aren’t prepared to deal with yet.
That’s not activist panic. That’s a Harvard professor describing the current state of affairs as genuinely ahead of our legal and social frameworks.
And the arms race dynamic is already baked in. If Nearby Glasses detects recording glasses, what stops someone from buying glasses with the LED disabled or obscured? If facial recognition gets added, what stops someone from using it silently? If the law says you can’t record people without consent, how do you enforce that against a device worn like normal eyewear in a crowd?
These aren’t hypotheticals. They’re the obvious next moves in a cat-and-mouse game that’s already started.
Why Norway Matters Even If You Don’t Live There
The instinctive reaction from the “innovation” crowd will be: one Nordic country overreacting, chilling a promising technology, etc. That’s the standard script for any privacy regulation and it’s usually wrong.
Here’s why this specific moment matters:
The technology is moving faster than the social contract. Smart glasses have been on the market long enough that the “wait and see” period is over. We’ve seen how they’re actually used — not in some theoretical best case, but in the real world, with real lawsuits, real investigations, and real people trying to hide from them. The data is in. The question is whether anyone in power is willing to act on it.
Norway is small enough to actually do something. A country of 5.5 million people can ban or restrict a product category within its borders. Meta and Snap can choose to withdraw rather than comply. That makes Norway a natural test case: if a mid-sized democracy says “no” to a category of surveillance technology, what happens? Does the industry adapt? Does the market find a workaround? Or does the rest of the world watch and realize this was a reasonable line to draw all along?
The facial recognition dimension changes everything. A camera you wear is one thing. A camera that can identify strangers in real time, cross-reference them against databases, and notify the wearer — that’s a different category of intrusion entirely. The 70+ advocacy groups that petitioned Meta understood this. Norway’s minister understood this. The stakes are similar to what’s playing out in the OpenAI copyright fight — where the question isn’t just what a company can build but what it can build using other people’s data without clear consent. The question is whether regulators in larger markets will catch up before the feature ships.
There’s a Filipino angle that doesn’t get talked about. We’re a country where public CCTV is already a contested issue, where data privacy law exists but enforcement capacity is limited, and where the gap between what’s legal and what’s practical is wide enough to drive a truck through. And the kind of personal-assistant privacy trade-off I wrote about in why the AI assistant everyone loves is a privacy red flag is already familiar territory here. Add wearable cameras that people can’t easily detect, and you’re layering a new surveillance vector onto an environment that’s already struggling with the old ones. The Norway debate isn’t abstract for countries that haven’t even started the conversation yet.
What Actually Needs to Happen
Banning the entire category is one option and Norway is exploring it. But there are narrower, more targeted approaches that might work better, and it’s worth sketching them out because “ban it” is not the only answer and may not be the best one:
- Mandatory, unambiguous recording indicators. A blinking LED the size of a pinhead is not enough. If a device can record people without them reasonably noticing, it needs a signal that ordinary humans can actually detect — and the signal needs to be verifiable, not something the manufacturer can claim meets the standard while engineering around it.
- Consent frameworks for public recording. If you’re going to record people in public and the device is inconspicuous, the consent model needs to account for that. Phone cameras already exist, but a phone is visibly a recording device when it’s pointed at someone. Glasses are not. The law needs to close that gap rather than pretending the existing phone-camera rules cover a fundamentally different use case. Something I explored in why AI detection is genuinely a spectrum, not a yes/no test.
- Real prohibition on covert facial recognition. The NameTag code WIRED found is the smoking gun here. A feature that identifies strangers without their knowledge, built into a device designed to be worn unnoticed, is precisely the kind of capability that needs to be blocked before it ships, not after it’s abused. Seventy advocacy groups asked for this. The fact that it’s still a live question is the problem.
- Data handling that actually matches the marketing. If a product is sold as “built for your privacy,” the footage it captures shouldn’t be reviewed by overseas contractors without clear, prominent disclosure that people actually notice. The lawsuit’s core claim — that there’s a gap between the privacy promise and the data practice — is not a frivolous one. It deserves a real answer, not a “our privacy policy says so” deflection. After something like the IDScan breach that exposed 150 million driver’s licenses, the stakes of who gets to see your data should be obvious to everyone.
- Independent auditing, not self-reporting. If a company says its face-blurring works consistently, that claim should be verifiable by someone other than the company’s own engineers. The BBC’s finding that blurring “didn’t consistently work” suggests the current system relies too heavily on trust and not enough on verification.
The Honest Take
I’m not someone who thinks all new technology is suspicious by default. I’ve spent my career in ICT, deploying tools that genuinely make work better. I’ve written about self-hosted photo backups, local LLM inference, and security tools that protect rather than surveil. I’m pro-technology. I’m also pro-privacy, and I don’t think those things are in opposition when the technology is designed honestly.
Smart glasses are a genuinely useful product in some contexts. Hands-free capture for accessibility, for documentation, for people who need it — there are legitimate use cases. The problem is that the current generation of products has been designed around a different priority: making surveillance frictionless, inconspicuous, and scalable. That design choice has consequences for everyone who isn’t the person wearing the glasses.
Meta’s marketing says “you’re in control of your data and content.” The fine print says overseas contractors may review your footage. The blinking LED says you’re being recorded, if you can see it. The facial recognition code says the company planned for a world where the glasses identify strangers automatically. And the reason Norway is considering a ban is that all of these things are true at the same time, and the public is increasingly uncomfortable with the combination.
That discomfort is not Luddism. It’s a rational response to a technology that collects data about non-users, by non-users, without their knowledge, and ships it through a pipeline that includes human reviewers overseas. When a Harvard professor says we’re not prepared for what’s happening, and a Nordic government says “we need to prevent this equipment from being used to monitor other people,” and a class-action lawsuit says the privacy marketing doesn’t match the data reality — that’s a pattern, not a noise.
The right response isn’t to pretend the concerns are overblown or that the technology will sort itself out. It’s to engage with what’s actually been built, how it’s actually being used, and what the people on the wrong end of the camera lens actually need to feel safe in public. Norway is starting that conversation. The rest of us should pay attention.