More than 70 organizations — the ACLU, EFF, Fight for the Future, Access Now, and a long list of international civil-liberties groups — just signed an open letter telling Meta to abandon facial recognition in its upcoming smart glasses. The letter, which landed this week, is short, surgical, and unusually angry. The ask is simple: do not ship a wearable camera that can silently identify strangers on the street.
Meta’s response, in so many words: we hear you, we’re thinking about it, and we’re probably going to ship it anyway.
That gap — between what 70 privacy experts are begging for and what Meta is actually planning — is the real story. Because once you understand why Meta won’t just say no, you understand what the next five years of consumer hardware are going to look like.
The quiet part nobody is saying out loud
Meta has spent a decade and well over $60 billion trying to build the device that comes after the smartphone. Quest headsets were the first bet. Ray-Ban Meta glasses were the second — and the first one that actually worked. They sold out. People wore them. The camera-on-your-face taboo cracked faster than anyone at Meta expected.
Now Meta is preparing the sequel: a next-generation pair with a real display, real compute, and — this is the part the privacy groups are fighting over — real-time computer vision that can identify anything the wearer is looking at. Including faces.
Why won’t Meta just carve out faces? Here is the honest answer, the one Meta’s PR team would never put on a slide: facial recognition is the feature. It is not an edge case. It is not a “power user” bonus. It is the thing that makes a pair of $800 glasses feel like magic instead of an expensive gadget. Walk into a conference, look at a name tag you cannot read, and the person’s LinkedIn profile appears in your field of view. Spot someone you met six months ago and forget their name — the glasses whisper it. That is the product. Without it, Meta ships yet another camera-on-your-face gadget and waits for Apple to eat its lunch.
Who gets hurt when the glasses ship
The 70 organizations are not worried about hypothetical harms. They have a list. Domestic abuse survivors whose ex-partners can now identify them in a crowd. Undocumented people who can be outed by a stranger’s glance. Activists, journalists, and sources whose cover depends on being unrecognizable in public. Sex workers. Trans people who have chosen not to be publicly identified with their old name or face. Women being followed by men who now have a tool that makes stalking trivial.
The argument Meta will make in response is that it already has “safeguards” — consent flows, little LEDs, terms of service. This is the same argument every surveillance product has made since CCTV. And it has never once survived contact with reality. Once a device is in the wild, it gets jailbroken, modded, or simply used exactly as designed against people who never agreed to anything.
Follow the money — and the regulatory map
Here is where it gets interesting. Meta cannot ship facial recognition in the EU. The AI Act prohibits real-time biometric identification in public spaces. It cannot ship it in Illinois without being sued into oblivion under BIPA. It probably cannot ship it in California, Texas, or Washington for much longer either. The regulatory noose is tightening everywhere except the two markets Meta cares about most: the rest of the United States, and India.
That is not a coincidence. India has no comprehensive biometric privacy law yet. Neither does most of the United States outside a handful of blue states. Meta’s calculation is that it can ship facial recognition in the places where regulators have not caught up, capture the market, and then negotiate from a position of strength when the laws finally arrive. It is the same playbook Uber ran on taxi regulation and Airbnb ran on short-term rental laws. Move first, apologize later, let the installed user base do the political lobbying for you.
The second-order effect nobody is pricing in
Every major hardware maker is watching this fight. Apple’s Vision Pro successor is in development. Samsung is shipping XR. Google’s Android XR platform is live. Chinese makers are building glasses that are already cheaper and lighter than anything in the West. The moment Meta ships facial recognition and does not get sued into bankruptcy, the norm flips. Every competitor adds it the next quarter because not having it becomes a spec-sheet disadvantage. And within 18 months, the default experience of walking down a public street is that half the people you pass can — technically — know your name, your employer, and your last five social media posts before you have finished making eye contact.
The privacy groups are not fighting one product. They are fighting the precedent. If Meta gets away with this, the question is no longer whether real-time facial recognition exists in consumer hardware. It is whether you are willing to wear a device that does not have it.
The verdict
Meta is not going to blink. The economics are too good, the regulatory gaps are too wide, and the competitive pressure from Apple and China is too real. The 70-organization letter is important not because it will change Meta’s mind — it will not — but because it draws the line in public before the product ships. Which means when the first harm lands, and it will, nobody in Menlo Park gets to say they did not see it coming.
The smart-glasses era is arriving. Whether it arrives with facial recognition baked in is being decided right now, in a letter most people will never read, by a company that has already done the math on who wins and who gets hurt.
