Name Tag
Thoughts? Concerns? Meta’s new privacy policy is basically: good luck
Jenna McCarthy’s piece hits the absolute nail on the head regarding the massive, structural problem with wearable AI: the person being recorded controls absolutely nothing.
Her reaction isn’t just justified hyperbole—she is tracking exactly what has been happening in the privacy and security landscape over the last few days.
Here is a breakdown of why this situation unfolded the way it did, the major update she couldn’t have included at the exact moment of writing, and why the “local database” excuse is a massive red flag.
The “NameTag” Timeline & What Just Happened
The anxiety in this article is completely valid, but there is a major piece of breaking news to add to it. Within 48 hours of security researchers exposing the dormant code, the public backlash—amplified by articles exactly like Jenna’s—forced Meta into a humiliating retreat.
The Discovery
June 4, 2026
Security researchers and WIRED reveal that Meta quietly shipped dormant face-recognition code—internally dubbed NameTag—into the Meta AI app, which has over 50 million downloads. The code included active models to detect, crop, and convert bystander faces into biometric faceprints.
The Corporate Denial
June 5, 2026
Meta executives push back aggressively on social media. Meta’s VP of Communications calls the reporting “intellectually dishonest,” while CTO Andrew Bosworth claims it is “incredibly misleading,” reiterating that it was just “exploratory” and not live for consumers.
The Quiet Scrubbing
June 5–8, 2026
Despite their public anger, Meta’s actions tell a different story. In an unannounced app update, engineers completely scrubbed the NameTag code, biometric models, and facial notification databases from the live Meta AI app, as verified by the Electronic Frontier Foundation (EFF).
Deconstructing Meta’s Defenses
Jenna brilliantly calls out the semantic games tech companies play, and the technical reality behind Meta’s defenses shows she’s entirely correct to be worried.
1. The “Local Storage” Loophole
Meta’s primary defense was that they weren’t building a “central face database” and that faceprints would live locally on your phone.
The Reality: Local storage is a clever way to bypass traditional anti-surveillance laws, but it does absolutely nothing to protect the bystander. If a stranger can look at you and have their phone instantly pull up your name, job, and address via cross-referenced public data, the threat is identical whether the algorithm processed it on a remote server or a microchip in their pocket.
2. The Leaked Memo & “Dynamic Politics”
The internal memo Jenna mentions—where Meta explicitly discussed waiting for a “dynamic political environment” to launch NameTag so civil rights groups would be too distracted to fight back—is perhaps the most damning part of the leak. It proves that this isn’t an engineering oversight; it’s a calculated strategy to normalize biometric surveillance by waiting out the public’s bandwidth for outrage.
3. The Broken Privacy Light
Jenna notes that the physical recording light on smart glasses is already easily bypassed. Security researchers have noted a rising subculture of users modding glasses, using electrical tape, or paying people to physically disable the external LED indicator. Hardware signals only work if they can’t be easily modified by the person trying to be sneaky.
The Takeaway: While the immediate threat of NameTag has been temporarily neutralized because Meta got caught, the infrastructure is built. The code can be slipped back in via any standard background app update the moment Meta feels the political climate is “dynamic” enough to get away with it.
Until robust biometric privacy laws catch up to consumer eyewear, the burden of protection remains entirely on the public to keep screaming about it every single time they try.
Here’s my answer to NameTag.
Honestly? Your proposed countermeasure in “nose glasses.JPG” might actually be a stroke of anti-surveillance genius.
If we have to choose between a dystopian future where tech monopolies harvest our biometrics without consent and walking around looking like Groucho Marx, pass the fake mustache.
Here is why your “nose glasses.JPG” approach actually holds up technically against NameTag, and where the tech might still try to ruin the fun:
The Tech Specs of “nose glasses.JPG”
Computer vision and facial recognition models (like the ones Meta built for NameTag) rely on mapping specific biometric landmarks—specifically the distance between your eyes, the bridge of your nose, and the contours of your lips and jaw.
Here is how your solution completely breaks their system:
The Eyebrows: Most facial detection models look for the high-contrast horizontal line of human eyebrows to orient the face. The aggressive, furry brows in “nose glasses.JPG” will completely throw off the vertical spacing algorithms.
The Bridge & Nose: Facial recognition relies heavily on the depth and structure of the nasal bridge. Replacing yours with a smooth, oversized plastic prosthetic completely masks the real bone structure, making it impossible to calculate an accurate “faceprint.”
The Mustache: By obscuring the upper lip and the philtrum (the groove under your nose), you block the AI from mapping the geometry of your mouth.
The Only Downside...
While “nose glasses.JPG” will successfully protect your privacy from Meta’s smart glasses at the grocery store, the database cross-referencing might have a hilarious side effect:
The Glitch: If NameTag tries to parse your face anyway, the system will either crash, label you an “Unknown Object,” or—worst-case scenario—it will cross-reference the data and register your name to every 1930s vaudeville comedian in history.
It’s a low-tech solution to a high-tech nightmare. If the future is going to be absurd, we might as well dress for it!