The UK Border Force is aiming to expand its use of AI to enhance the searching and screening of freight at UK seaports.

The agency is seeking to automate the detection of anomalies in x-ray images, as manual analysis is both time-consuming and prone to error. AI systems could be employed to better balance security with the smooth movement of legitimate goods, the Border Post hopes.

AI-driven automation may not only speed up screening times but also increase the volume of goods processed. Additionally, it could free up officers to focus on critical tasks, reduce false alarms and minimise unnecessary secondary inspections.

For these ambitions, Border Force turning to the Accelerated Capability Environment (ACE) for support in organising and analysing its extensive x-ray image database, which had previously been stored inconsistently across multiple data structures.

ACE was initially tasked with creating a fully indexed and standardised repository of x-ray images and associated data – making it easily accessible to analysts and suitable for AI algorithm development.

Six ACE suppliers – Faculty, Leonardo, Ploygeist, Roke, Symetrica, and Zaizi – collaborated to explore AI and machine-learning methodologies for anomaly detection. Using the newly indexed data, they developed three use cases: vector integrity, pattern recognition and high-density material detection.

Recommended reading

• UK Policing AI and Automation Called Out Over Racial Bias
• Starmer Announces £75M for UK Border Security Tech
• Civil Rights Alarm Bells Raised over PM’s Surveillance Plans
• UK Police Use of Facial Recognition Found to be ‘Unethical’

These were presented to Border Force at a demonstration day, where the solutions were then taken forward for trials in a controlled operational environment.

Following an analysis of Border Force’s data, systems and applications, which saw their date consolidated into a unified structured format, trials were conducted with Border Force staff across multiple ports to evaluate the AI anomaly detection system.

The success of these commissions has led to a request for further development, forming the foundation of a business case for wider AI implementation across Border Force operations.