Full Report
Customs and Border Protection has called for tech companies to pitch real-time face recognition technology that can capture everyone in a vehicle—not just those in the front seats.
Analysis Summary
# Main Topic
U.S. Customs and Border Protection (CBP) is seeking proposals from technology vendors for real-time facial recognition technology capable of comprehensively scanning all occupants within a vehicle at land border crossings, specifically targeting passengers in rear rows, to match them against travel documents.
## Key Points
- CBP issued a Request for Information (RIF) to procure technology to expand existing air/sea/pedestrian facial recognition capabilities into the "land vehicle environment."
- The goal is to achieve "100% of vehicle passengers" capture, augmenting current systems that struggle due to vehicle geometry and passenger positioning (e.g., those in the second or third row).
- The current system performs one-to-one facial recognition, matching a real-time capture against identity documents provided by the traveler.
- Recent testing at the Anzalduas port of entry (Mexico border near McAllen, TX, late 2021 to early 2022) showed system limitations: cameras captured everyone in the car only 76% of the time, and of those captured, only 81% met the "validation requirements" for reliable matching.
- The primary risk highlighted by critics is the failure of the system to recognize a match when inputs are substandard.
## Threat Actors
- **Threat Actor:** US Customs and Border Protection (CBP), an agency under the Department of Homeland Security (DHS).
- **Motivation:** Securing borders and biometrically confirming entries into the United States by adding confirmed data to traveler records.
- **Attribution:** Explicitly identified as CBP seeking vendor solutions via a federal register posting.
## TTPs
- **Targeted Environment:** Land vehicle ports of entry (POEs).
- **Technique Sought:** Real-time facial recognition augmentation to capture images of passengers obscured or positioned in the rear rows of vehicles.
- **Current Operational Mode:** One-to-one facial recognition matching against travel documents and existing government holdings.
- **Challenges Identified:** Human behavior, multiple passenger vehicle rows, and environmental obstacles hinder 100% capture rates.
## Affected Systems
- **Primary Target Systems:** Existing CBP real-time face recognition systems currently operating in air, sea, and pedestrian environments.
- **Scope of Impact:** Expansion of biometric surveillance into the land vehicle environment at U.S. POEs.
- **Affected Individuals:** All individuals traveling through land ports of entry in vehicles, including those in rear seating positions.
## Mitigations
*(Note: As the article details a government procurement effort rather than an external threat attack, suggested mitigations focus on oversight and privacy protection concerning the implemented surveillance technology.)*
- **Transparency and Oversight:** Public review of the RIF and associated tests conducted at ports like Anzalduas is crucial for oversight.
- **Error Rate Minimization:** Vendors must propose solutions that significantly improve the capture rate (currently 76%) and the validation match rate (currently 81%) observed in pilot tests.
- **Privacy Protection Protocols:** Tech companies pitching solutions must address concerns raised by privacy advocates regarding bulk collection of passenger biometrics, which could imply a shift toward mass surveillance rather than targeted confirmation.
- **Document Handling:** Strict protocols must be enforced for securely handling and destroying images that do not result in an official "biometrically confirmed entry."
## Conclusion
CBP is aggressively pursuing technological upgrades to implement comprehensive biometric scanning of all vehicle occupants at land borders. The current operational technology demonstrates significant technical hurdles, particularly concerning image capture quality and reliable matching across varied passenger seating arrangements. The adoption of this enhanced technology implies a major expansion of government surveillance capabilities at land entry points, necessitating scrutiny regarding accuracy rates and privacy safeguards.