Exclusive: Meta Tests Police/Military Surveillance Facial Recognition for Smart Glasses, WIRED Has Learned

Exclusive reporting from WIRED confirms Meta is testing facial recognition software built by Rank One Computing, a vendor that supplies surveillance tools to U.S. law enforcement and the U.S. military, as the tech giant explores adding the capability to its consumer smart glasses line.

Full details of the partnership are outlined in a software license obtained by WIRED. The agreement was issued by Rank One, a Denver-based firm that pulls roughly 80% of its total revenue from government clients, and is linked to a test build of the Meta AI app that powers Meta’s Ray-Ban and Oakley smart glasses.

Rank One’s facial recognition tech is already widely deployed across U.S. federal law enforcement and military agencies: The U.S. Marshals Service uses it to verify inmate identities during transit without requiring fingerprint scans, while the Naval Criminal Investigative Service (the Navy’s law enforcement division) purchased Rank One’s video surveillance tool ROC Watch. Under a federal research contract, Rank One also developed long-range facial recognition for U.S. Special Operations Command, advertising its software can accurately identify a face from as far as one kilometer away. Local police departments across the U.S. also rely on Rank One’s algorithms, which are embedded in third-party surveillance tools sold to departments by other vendors.

This license marks the first public evidence of a business relationship between Meta and Rank One, and offers a rare look at the technology Meta is evaluating for a mass-market consumer device. It also underscores how blurred the line has become between surveillance technology built for military and law enforcement use, and consumer products sold to the general public: increasingly, the same companies and core core algorithms serve both sectors.

The license grants Meta rights to use Rank One’s facial recognition alongside the firm’s liveness detection tool, which confirms a camera is capturing a live person rather than a photo or mask. The agreement supports up to 10 million facial templates and remains active under its terms. Code reviewed by WIRED shows traces of Rank One’s integration—including code routines that load the license and initialize Rank One’s software—remained dormant in a version of Meta’s app shipped to millions of consumers earlier this month, alongside Meta’s own in-house facial recognition system.

None of the facial recognition tools tied to Meta’s smart glasses have ever been activated for end users. Meta removed all related code from the app entirely on June 5, one day after WIRED revealed the company had quietly built an unlaunched facial recognition system (internally named NameTag) into the Meta AI app, the companion software for its smart glasses that has been downloaded to more than 50 million consumer phones. The internal system was dormant and inaccessible to regular users.

Meta has offered almost no comment on the arrangement, declining to answer WIRED’s questions about its relationship with Rank One. The company would not confirm why it licensed the software, when the partnership began, or if it remains ongoing. Rank One also declined to comment for this story.

| Got a confidential tip? |

|---|

| Are you a current or former Meta employee with insight into the company’s emerging technologies? We want to hear from you. Using a non-work phone or computer, contact a reporter securely via Signal at dmehro.89 or dell.3030. |

Founded in 2015, Rank One Computing was launched by a team of engineers who previously built facial recognition systems at non-profit research institute Noblis, work that included evaluating algorithms for a U.S. intelligence research agency. The company went public on the Nasdaq exchange in February 2024.

Rank One’s leadership is drawn almost entirely from senior ranks of U.S. law enforcement and intelligence. Its CEO, B. Scott Swann, previously led the FBI division that manages the bureau’s biometric databases. Its board includes a former CIA deputy director for science and technology, a former head of the FBI’s science and technology branch, and a former Pentagon official who built a multi-billion-dollar special capabilities office for the department.

“There’s a long history of military technologies eventually being adapted for consumer products,” said Joseph Jerome, a former Meta Reality Labs policy official. “Arguably, that’s the origin story of the internet itself.”

Rank One’s technology is already deployed in a range of high-profile public settings. The U.S. Marshals Service has used a biometric identification kit built on Rank One’s tech since 2021. In 2024 comments, Rank One’s CEO confirmed dozens of West Virginia public schools use the software to screen visitor faces at entrances against the state’s sex offender registry. Rank One’s algorithm is also bundled into products from South Carolina-based firm DataWorks Plus, and LexisNexis’s Lumen platform, which allows police to run facial searches across state, regional, and FBI national investigative databases.

Like most commercial facial recognition systems, Rank One’s tool does not perform equally across demographic groups. Testing by the U.S. National Institute of Standards and Technology (NIST) found one version of Rank One’s algorithm produced false matches at drastically different rates based on a person’s sex and country of birth (which NIST uses as a proxy for race). Error rates were lowest for people born in Eastern Europe, and consistently higher for women than men.

There is almost no federal regulation governing facial recognition use in the U.S. Eric Null, director of the Privacy and Data Project at the Center for Democracy and Technology, notes that while many states require police to obtain a warrant before accessing facial recognition data, and more states add biometric protections to consumer privacy laws each year, rules for consumer-facing companies remain sparse.

“But consumer-facing companies clearly crave access to high-powered facial recognition technology,” Null said. “And without proper checks, the risks of this tech becoming a common consumer product are significant and largely unbounded.”