AI Co-Pilot Could Help Make Digital Forensics More Efficient

Digital evidence plays a central role in almost every criminal investigation. As law enforcement agencies face growing volumes of digital data to analyse, the availability of digital forensics experts remains limited. This creates a structural workload and increasing reliance on specialist expertise, putting pressure on the efficiency and turnaround time of investigations.

Hansken, a digital forensic platform developed by the Netherlands Forensic Institute (NFI), makes large datasets of digital evidence accessible to law enforcement agencies. In practice, however, effective use of the platform still depends heavily on specialist knowledge, meaning digital forensics experts are regularly required to carry out routine analysis tasks.

A new KIEM project is investigating whether an existing AI co-pilot can support the use of Hansken in such a way that non-specialists can carry out the analysis of digital evidence more independently. The project focuses on whether AI can help reduce the need for expert intervention and shorten analysis times, while maintaining or improving the quality of digital findings.

As part of controlled experiments in a realistic working environment, approximately 60 professionals will perform analysis tasks both with and without AI support, using a cross-over study design. The researchers will combine quantitative measurements, such as task completion times and the number of expert interventions, with qualitative assessments, including expert evaluations of the results and user experiences. This will provide a systematic assessment of the potential time- and labour-saving impact of AI.

The project specifically focuses on testing an existing, readily deployable AI application in a realistic operational setting. In doing so, it aligns closely with the objectives of the KIEM funding scheme, which aims to explore the practical potential of AI applications that can reduce workloads.

The findings will provide valuable insight into the extent to which digital evidence analysis can be scaled up with less reliance on specialist expertise, as well as the conditions required for the responsible use of AI, including reliability, transparency and human oversight. The results will provide an empirical foundation for further research and potential scaling across the digital investigation chain.

Source: SIA Projecten