Opinion: The Semantic Shell Game: Meta’s War on the Truth About NameTag

AI-generated image · US National Wire
By pivoting between claims that its face-recognition tech 'doesn't exist' and detailed descriptions of its utility, Meta isn't just confused—it's attempting to gaslight the public on its surveillance reach.
Meta is currently engaged in a masterclass of corporate obfuscation, attempting to rewrite the definition of existence to shield itself from accountability. The target of this semantic warfare is NameTag, a face-recognition system developed for the company's smart glasses.
As Wired first reported, the tension began when the outlet revealed that Meta had integrated robust, albeit inactive, code for NameTag into the Meta AI companion app. This app, used by tens of millions of people, contained the system's core components as early as January. By May, core components of the code were present. Despite this, Andy Stone, Meta’s vice president of communications, took to X to claim the feature "doesn't exist," subsequently accusing Wired of "intellectually dishonest" reporting and "advocacy-driven click bait."
This narrative of non-existence is not just a corporate blunder; it is a calculated contradiction. While Stone claimed the company could not answer questions about how the system works because it does not exist, Meta CTO Andrew “Boz” Bosworth spent several minutes on a July 8 episode of *The Most Interesting Thing in AI* podcast describing the feature in detail. Bosworth explained to host Nicholas Thompson that NameTag would allow a user to remember someone they met in person while wearing glasses, stating, "this is a person you’ve met before. Here’s their name. They’re right in front of you."
When Wired pressed Meta on the disconnect between Stone’s denial and Bosworth’s detailed explanation, the company retreated into linguistic gymnastics. Spokesperson Ryan Daniels told Wired there was "no contradiction," specifically bolding and underlining the word "would" in Bosworth's statement to argue that the CTO was speaking conditionally.
However, the technical reality contradicts the corporate rhetoric. Wired reports that Meta had been building NameTag since early 2025, which included licensing face-recognition software and assembling a full detection-and-matching pipeline. Furthermore, a researcher known as Buchodi, reviewing the code at Wired's request, successfully used the NameTag system to recognize a photograph of philosopher Michel Foucault.
Perhaps most concerning is Meta's insistence on a distinction that serves its own image. Bosworth asserted that NameTag would not rely on a "central database," a claim Wired notes was unnecessary as neither the podcast host nor the outlet had suggested such a thing. Wired's analysis found that NameTag converts captured faces into "faceprints"—unique numerical signatures. These could then be compared against a database stored on the user's device, which is populated by Meta’s own servers.
By claiming the tech doesn't "exist" while the system's code sat on millions of devices and describing its future utility, Meta is attempting to avoid a public reckoning over its surveillance capabilities. This isn't a misunderstanding of terminology; it is a strategic effort to minimize the perceived scale of their facial recognition ambitions.

