Request for Information (RFI) - Artificial Intelligence for Image Adjudication

The U.S. Department of Homeland Security, Customs and Border Protection is seeking information from industry on how to best develop Artificial Intelligence (AI) and Machine Learning (ML) models for image adjudication in Non-Intrusive Inspection (NII) systems. The goal is to assist CBP employees in making better decisions faster, reviewing and synthesizing data, and improving efficiency. The AI/ML models should be able to operate on large-scale NII imagery and associated metadata, and be adaptable to CBP's operational needs and constraints. The models should also be able to be integrated as drop-in components within existing CBP architectures. Some key points to consider include: • The ability to detect anomalies and identify contraband or modifications • The ability to operate on various image formats, including those that conform to the World Customs Organization (WCO) Unified File Format (UFF) • The ability to be integrated with existing CBP data platforms and pipelines • The ability to be trained, verified, and deployed in a way that minimizes technical risk and life cycle costs • The ability to be adapted to different operational settings and environments • The ability to provide automated identification of anomalies, identification of conveyance modifications, manifest validation, contraband and narcotic identification, and verification of commodities.