CBP Pushes for Tech to Capture Faces of Everyone Crossing the Border by Car

U.S. Customs and Border Protection (CBP) is seeking proposals from tech vendors for a facial recognition system that would capture and identify every person inside a vehicle crossing the border, including those in back seats. The goal is to match each face to existing travel or identity documents, according to a Request for Information (RFI) posted last week.

CBP already uses facial recognition at air, sea, and land pedestrian ports of entry, but extending the system to vehicles has proven technically challenging, as Wired explained in a sprawling report on Tuesday . Environmental obstacles, seating arrangements, and human behavior hinder the agency's ability to consistently capture usable images of all passengers.

A 152-day test at the Anzalduas border crossing in Texas, cited by the report as an example, revealed that the current system captured images of all occupants in only 76% of vehicles. Of those, only 81% met facial validation requirements to match with identity documents.

"The current system is one-to-one facial recognition," said Dave Maass, director of investigations at the Electronic Frontier Foundation to Wired. "The risk is the system failing to recognize that someone matches their own documents." This differs from one-to-many facial recognition systems, often used in policing, which carry the risk of false matches.

CBP has not specified whether the capture issues stem from the image-gathering cameras or the software performing the matching, as Maass noted, "We don't know what racial disparities, gender disparities, etc., come up with these systems."

CBP's call for enhanced surveillance technology follows the recent disclosure that Immigration and Customs Enforcement (ICE) has contracted Palantir Technologies to build a $30 million platform called ImmigrationOS, designed to enable near real-time tracking and analysis of undocumented migrants, particularly those who self-deport.

According to ICE, the platform will integrate biographic, biometric, and behavioral data from various sources to identify individuals based on factors including country of origin, visa status, tattoos, and vehicle movements. The ICE contract was awarded without competitive bidding and justified as essential to addressing threats from transnational criminal organizations.

Palantir, founded by Peter Thiel, has longstanding ties to U.S. immigration enforcement. ImmigrationOS builds on the case management system Palantir has provided to ICE since 2014, which already integrates a wide range of personal data, including from covert tracking devices and license plate readers.

Though driven by the current administration's enforcement goals, Maass emphasized that CBP's surveillance expansion is not something unique to the current Trump administration.:

"CBP surveillance strategy carries over from administration to administration—it always falls short, it always has vendor issues and contracting issues and waste issues and abuse issues. What changes is often the rhetoric and the theater around it"

Originally published on Latin Times