U.S. Customs and Border Protection is seeking industry input on the use of artificial intelligence and machine learning to enhance image adjudication across its Non-Intrusive Inspection (NII) systems, signaling a major step toward modernizing border screening operations.
The objective of this effort is to integrate AI/ML capabilities into existing inspection workflows to improve the speed, accuracy, and efficiency of analyzing large-scale X-ray imagery. CBP currently processes high volumes of imaging data from cargo and passenger vehicle scanning systems, and the growing scale of operations is placing increasing pressure on human analysts.
Scope centers on the development of algorithmic tools that can assist with anomaly detection, contraband identification, conveyance modification detection, and manifest validation. Rather than replacing existing systems, CBP is specifically seeking modular AI/ML components that can be integrated into its current architecture, supporting a “human-in-the-loop” model where officers retain decision authority.
A key focus is scalability and interoperability. Solutions must operate across diverse imaging formats and environments, handle inconsistent or incomplete metadata, and process large volumes of data in near real time. The effort also emphasizes open architecture approaches, enabling integration with existing CBP platforms without reliance on proprietary data pipelines or systems.
The initiative reflects CBP’s broader strategy to expand NII scanning coverage while maintaining operational efficiency and the flow of legitimate trade and travel. By leveraging AI/ML, the agency aims to increase throughput, improve threat detection, and optimize resource allocation across ports of entry.
This opportunity will likely attract firms specializing in computer vision, AI/ML model development, image analytics, and high-performance data processing, particularly those with experience in security screening, industrial imaging, or large-scale inspection systems.
