In an effort to enhance its border security operations, the U.S. Customs and Border Protection (CBP) is actively seeking input from industry experts on the development of artificial intelligence and machine learning (AI/ML) tools. These tools aim to assist in analyzing images captured at border checkpoints through the agency’s non-intrusive inspection (NII) systems.
On April 30, CBP published a request for information (RFI) on SAM.gov. The agency’s goal is to “reduce the average NII examination processing time” while simultaneously increasing the volume of scanned vehicles. The overarching aim is to enhance security outcomes and ease the workload for officers, all without hindering commerce.
The CBP Office of Field Operations, via its NII Program, is specifically concentrating on large-scale imaging systems that produce substantial amounts of X-ray image data and metadata from both private and commercial vehicles. These systems are currently implemented across various locations, including airports, seaports, and land border crossings, with plans for further expansion in both scale and performance, according to CBP.
As the amount of scan data grows, it has surpassed the capacity of human analysis, prompting a need for AI/ML tools that can aid officers in detecting anomalies, identifying threats, and making informed decisions.
CBP highlighted that “frontline personnel remain the most critical asset,” underscoring the role of AI/ML as a supportive mechanism within a human-centric operational model.
Through this RFI, CBP is interested in acquiring modular algorithmic solutions rather than comprehensive data platforms. These solutions should seamlessly integrate with existing CBP-owned systems and architectures. The agency emphasizes that it is not seeking proprietary data pipelines or case management systems, but rather components that can work effectively within its current framework.
The RFI identifies several key capability areas of interest, including anomaly detection, identification of modifications to conveyances, validation of manifests, and the detection of contraband such as narcotics, weapons, and currency.
Additionally, CBP is looking for tools that can segment images, highlight areas of concern, and assign risk scores to facilitate officer review.
Responses from the industry are requested by May 30.
