5 Reasons Why Disparate Data Blocks AI Investigation Agents
Banks are pointing agents at systems that were designed for individual business processes — not for connected, iterative investigations. Michael Shearer, former Managing Director within HSBC Financial Crime Threat Mitigation, explains where this approach fails first, and why agents need a governed knowledge layer instead.
Inside the paper
- The link your AI never finds: the hidden connection a name-only search misses.
- When “no match” becomes a hallucination: why AI invents plausible answers.
- Business context AI doesn't understand: how payment semantics change the conclusion.
- The hidden cost of agentic investigations: why uncontrolled queries become expensive.
- Six questions before production: a practical readiness checklist.
“AI agents need more than access to data. They need a governed map of the entities, relationships, and business meaning they should follow”
— Michael Shearer, Former Managing Director, HSBC Financial Crime Threat Mitigation
Two decades leading intelligence and investigation teams — public sector, then HSBC.
Read Michael Shearer's Paper
Enter your business email to download the paper.
By submitting, you agree to receive this resource and closely related communications from DataWalk. See our Privacy Policy.
Introduction
Agentic AI has extraordinary abilities to analyze language, make connections, and compose answers to specific questions. It can also summarize presented facts, generating accessible natural language explanations for observed behaviour.
Assembling these complex investigative jigsaw puzzles piece by piece through repeatedly querying raw data sources may ultimately be satisfactory, however it has the potential to be uncontrollably expensive and introduces new risks to the investigative process.
In this paper we will explore five reasons why disparate data blocks AI investigation agents.…
