The Washington Post: “FBI, ICE find state driver’s license photos are a gold mine for facial-recognition searches.”

by Georgina Escobar


Federal Bureau of Investigation and Immigration Customs Enforcement agents use state department of motor vehicle databases for facial-recognition purposes, reveal newly released documents. These records, obtained by researchers with Georgetown Law’s Center on Privacy and Technology and shared with The Washington Post, contain thousands of facial-recognition requests, internal documents, and emails over the past five years.The Washington Post reports that “DMV records contain the photos of a vast majority of a state’s residents, most of whom have never been charged with a crime,” affecting millions of Americans whose photos are being used without their knowledge. Lawmakers across the aisle have criticized the technology as a “dangerous, pervasive and error-prone surveillance tool.” House Oversight Committee Chairman, Rep. Elijah E. Cummings (D-MD), and the committee’s ranking Republican, Rep. Jim Jordan (R-OH), publicly expressed their concerns about the lack of consent surrounding the use of DMV images.

A hearing held last Wednesday before the House Homeland Security Committee included news reports detailing Department of Homeland Security’s apparent mining of US motorists’ photographs in multiple states without a judge’s approval, as well as a recent data breach of Customs and Border Protection’s data. The main focus of Wednesday’s hearing was the question of whether CBP, in particular, has exceeded its legal authority to utilize face recognition. Under current law, CBP is only permitted to collect the biometric information of foreign nationals through what is known as the Biometric Exit Program. And while the ultimate question of whether CBP is operating within the scope of the Biometric Exit Program remained unresolved, the hearing raised concerns about the unreliability of face-recognition technology. In a telling example, a recent study conducted by the American Civil Liberties Union (ACLU) last year revealed that Amazon’s face recognition tool “Rekognition” falsely matched 28 members of Congress with criminal mugshots.

The FBI, however, claims that its system is “86 percent accurate at finding the right person if a search is able to generate a list of 50 possible matches,” according to the Government Accountability Office. However, the FBI has not tested its system’s accuracy under conditions that are closer to normal, including when a facial search returns only a few possible matches, according to The Washington Post. Some cities, including San Francisco, CA , and Somerville, MA, have responded to this controversy by banning their police and public agencies from using facial-recognition software, citing concerns about “governmental overreach and a breach of public trust.” The reliability and accountability of the government’s facial recognition systems are expected to remain “hotly debated in Washington.”