Revving Up Surveillance: DHS Eyes In-Motion Face Recognition at the Border - A Privacy Pandora's Box?

The Department of Homeland Security (DHS) has set its sights on an ambitious new project: deploying facial recognition scans on drivers and passengers as they approach the southern border. As detailed in a DHS document obtained by The Intercept, the initiative aims to capture the facial likenesses of travelers while their vehicles are still in motion, even before they pull up to a checkpoint. The DHS Science and Technology Directorate has put out a call for technology solutions that can gather biometric data, specifically facial recognition, from occupants of speeding vehicles. The goal is to enable checkpoint officials to swiftly determine if the occupants pose a threat, thereby expediting entry for compliant travelers without requiring them to leave their vehicles. Despite the patchy performance record and persistent shortcomings of facial recognition technology, DHS seems hell-bent on moving forward with this project. Vendors presenting their wares to DHS may be tapped to participate in additional testing. DHS and Customs and Border Protection (CBP) have so far refrained from commenting on this matter. Dave Maass, investigations director at the Electronic Frontier Foundation, has expressed concerns about the potential intrusiveness of facial recognition in the realm of personal privacy. Should law enforcement agencies gain the capability to capture facial recognition from moving vehicles, they would be empowered to track not only the whereabouts of vehicles but also the identities of drivers and their passengers. This, Maass argues, would open up a whole new frontier of privacy invasion.CBP has maintained that it has a legislative mandate to broaden biometric identity checks across land, air, and sea. While facial recognition cameras are already operational in major American airports, passengers can still opt out of this process for the time being. However, DHS has encountered challenges in identifying drivers remotely, as noted in a 2024 report by the DHS Office of Inspector General. A 2022 DHS postmortem report on the Anzalduas test revealed that drivers and passengers were successfully photographed only approximately three-quarters of the time, and only about 80 percent of these images were deemed usable. The 2022 report also highlights the inherent difficulties in using facial recognition technology effectively in outdoor border crossings, where factors such as windshields, reflections, hats, sunglasses, shadows, weather conditions, Covid masks, sun visors, and other real-world obstructions often interfere with image quality. As DHS turns to the surveillance industry for assistance, a cautionary tale can be gleaned from the 2022 report. In 2019, Perceptics, a company that provides CBP with license plate-scanning technology for the same checkpoints, fell victim to a hack that exposed unauthorized copies of traveler PII. These copies had been illicitly transferred to Perceptics’ corporate servers. The report does not offer further information on the data protection and insider threat security controls that were implemented or neglected, leaving open questions about the potential risks of