Meta Is Building a Face Scanner. You Didn't Vote for That. Neither Did Your Neighbours.
- Michael Routhier
- 5 days ago
- 7 min read
I want you to picture something.
You're sitting in a coffee shop. You're waiting for a friend. A stranger walks in wearing a pair of Ray-Ban glasses; nice looking, normal. They glance in your direction.
And their glasses know your name.
Not because you told them. Not because you agreed to anything. Because their glasses looked at your face, converted it into a biometric signature, and matched it against a database. In under a second. Without a sound.
That is not a hypothetical. That is the feature Meta has been quietly building inside an app that has already been downloaded to over 50 million phones.
They call it NameTag.
What Wired Actually Found
On June 4th, 2026, Wired published a technical investigation that should have been the lead story on every news channel in the world.
Buried inside Meta's AI companion app, the one you need to use the Ray-Ban and Oakley smart glasses, Wired found fully functional facial recognition code. Not theoretical. Not in a lab. On your phone. Distributed quietly across multiple app updates, starting as early as January 2026, months before Meta breathed a word of it publicly.
The feature works using three AI models already installed on your device. One detects faces. One crops and repositions the image. One converts that face into a unique biometric signature, a faceprint, which gets checked against a database stored on the wearer's phone. Faces that are recognized trigger an alert. Faces that aren't recognized? They get cropped, indexed, and saved to a folder marked "pending."
Pending. As in; we'll figure out what to do with these faces later.
When Wired asked about this, Meta's spokesperson said the company is merely "exploring" these types of features, and that what Wired found is "just evidence of that exploration."
Just evidence of that exploration.
They shipped the code to 50 million phones. They built three functional AI models. They created a folder for storing strangers' faces. And they want you to understand that as exploration.
The "It Only Uses Your Contacts" Reassurance, And Why It's Meaningless
I know what some people are going to say. They're going to say: "But Michael, it only identifies people in your contacts. People you already know. What's the harm in that?"
Let me explain exactly what's wrong with that framing.
First - who controls what "your contacts" means? Your contacts list isn't a static, private vault. It's a database that Meta has had access to for years. Every time you've given an app permission to access your contacts, every time someone who has your number uploads their contacts to Facebook or Instagram or WhatsApp, your information has been flowing into Meta's ecosystem. The boundary between "people I know" and "people Meta's system knows about" is not the clean line the reassurance implies.
Second - the architecture doesn't stay at "contacts". The system as built doesn't identify people; it creates faceprints of every face the glasses capture, stores the unrecognized ones in a "pending" folder, and checks them against a database that is configured to receive updates from Meta's servers. That is not a contacts lookup tool. That is the foundational infrastructure of a universal face database. The "contacts" framing is the first use case. It is not the last.
Third - and this is the one I need you to really hear, what happens when this data exists and a government comes asking for it?
Because that question has already been answered. Not by speculation. By history.
Meta has complied with government data requests tens of thousands of times. Every major tech platform has. When a law enforcement agency presents a legal demand, these companies hand over what they have. That is not a conspiracy theory, it is a matter of public record in their own transparency reports.
Now ask yourself, what is in that data?
Right now, it's messages and posts and location history. It's already invasive. But add faceprints, unique biometric signatures tied to identities, generated every time someone wearing these glasses looks in your direction and you have something categorically different. You have a database of where specific human faces were, when, and in whose company.
That is not a contact list. That is a surveillance log of everyone who has ever been seen by a Meta glasses wearer. And you never agreed to be in it.
The Harvard Students Already Showed You This Movie
In 2024, two Harvard students demonstrated exactly where this ends up.
They took a pair of Meta Ray-Ban smart glasses; the first generation, without NameTag, without any built-in facial recognition, connected them to external facial recognition software, and walked around Harvard's campus.
They could identify strangers' names, ages, home addresses, and phone numbers in real time. Not with any special access. Not with government tools. With off-the-shelf software and a pair of $300 glasses.
That was 2024. Before NameTag. Before Meta built three dedicated AI models into their app and started storing biometric data on users' devices.
The ACLU and 75 organizations sent a letter directly to Mark Zuckerberg calling this technology "a red line society must not cross". They warned it would allow anyone wearing the glasses to identify people at protests, at medical clinics, at religious services, and link those identities to databases containing health information, political affiliations, relationships, and home addresses.
Seventy-five organizations. The American Civil Liberties Union. A letter to Zuckerberg.
Meta kept building.
This Is What "Thoughtful Approach" Looks Like
In April 2026, Meta said publicly that if they were to utilize face recognition, they wouldn't roll it out "without first taking a very thoughtful approach".
In January 2026, three months earlier, the core components of NameTag were already being shipped to millions of phones.
I want you to hold both of those facts in your mind at the same time. The public statement about caution. The private action already underway. And ask yourself, which one is the real Meta?
This is the same company that, in 2023, was fined $1.3 billion by the European Union for illegally transferring European users' personal data to the United States. The largest privacy fine in history at the time. The same company that paid $650 million to settle a class action lawsuit in Illinois for collecting biometric facial data without user consent.
Not a cautious company. Not a company with a track record of earning trust on privacy. A company that has been fined and sued for exactly this kind of behaviour, and kept going.
The Virtuous Machine Question
I ask it about every piece of technology we cover on this platform; What is this actually for?
Not what Meta says it's for. Not the press release version. What is the system; NameTag, the pending folder, the faceprint database, the 50 million phones it's already installed on, actually for?
It is for the elimination of anonymity in public space.
That is the end state of this technology, whether Meta intends it or not, whether the feature launches with "contacts only" framing or not. Once faceprints are being generated for every face that passes in front of these glasses, and stored, and updated from Meta's servers, and accessible to legal requests, the concept of moving through the world without being identified, logged, and connected to a database is gone.
Not for criminals. Not just for activists or protesters or people with reasons to be afraid. For everyone. For your daughter walking to school. For your neighbour at their doctor's appointment. For you, sitting in a coffee shop, not doing anything wrong, just existing in a public space without having agreed to be scanned.
The ACLU called it destroying "the expectation of privacy or anonymity in public" and said it would "chill free expression for everyone".
They're right. And they said it four months ago. Meta kept shipping the code.
What You Can Do Right Now
I'll be direct, because that's what you come here for.
One: Do not buy Meta smart glasses. Not the Ray-Ban. Not the Oakley. Not any future version until this is resolved with legally binding protections, not press statements. Every pair sold expands the network of cameras that can feed this system.
Two: Check your phone for the Meta AI app. If it's installed and you don't use it, delete it. The facial recognition code is sitting on your device whether you use the glasses or not.
Three: Contact your elected representative and ask one question; What is our country's legal framework for preventing the collection of biometric data of citizens without their consent in public spaces? If they can't answer that clearly, now you know what needs to change.
Four: Support the Electronic Privacy Information Center (EPIC) at epic.org and the ACLU at aclu.org. These are the organizations doing the legal and advocacy work to push regulators to act. They need visibility, support, and public pressure behind them.
Marcus Aurelius wrote; "If it is not right, do not do it. If it is not true, do not say it."
Meta said in April they hadn't decided whether to release NameTag. They had been shipping its code since January.
That is not a thoughtful approach.
That is not transparency.
And it is not something any of us agreed to.
Michael Routhier is the founder of Tech 4 Grown-Ups, a platform dedicated to honest, unfiltered digital literacy for adults 55 and over. The Virtuous Machine is a series exploring the ethics, power, and human cost of artificial intelligence. Find everything at tech4grownups.com.
Additional sources: ACLU Open Letter to Mark Zuckerberg, April 2026 | Electronic Privacy Information Center (EPIC), February 2026 | New York Times, February 2026 | Forbes / Harvard student investigation, October 2024
Comments