Applied AI for Due Diligence Research: Whitepaper

We’ve published a new Threat.Digital whitepaper, Artificial Intelligence and the Future of Due Diligence, on how large language models are already changing investigative research. It looks at practical use cases we see in client work: faster scoping of assignments, earlier identification of issues across large data sets, and continuous monitoring that reduces noise in alerts by more than 94% while keeping full recall.

The paper also spends time on failure modes. It explains how LLMs can hallucinate facts, misread correct data, or produce accurate summaries of weak sources, and why source review, context, and non-digital intelligence still need experienced investigators.

Finally, it outlines how firms are beginning to use agentic AI to explore leads and related parties, with humans directing scope and making final calls.

You can read the full whitepaper here.