Why the old model costs so much
Traditional due diligence is expensive because a human analyst spends time searching across fragmented systems, resolving names, reading sources, comparing contradictions, asking for missing documents, and writing a narrative report.
The report is then delivered into a transaction and often discarded. The next transaction starts again.
What AI changes
AI reduces the marginal cost of search, summarisation, source comparison, contradiction detection, and report drafting. It can read more sources than a human analyst can read manually and produce a structured first pass quickly.
But AI does not remove the need for provenance. Compliance buyers still need to know what source was used, how strong the match was, who reviewed it, what the subject contributed, and whether the evidence is current.
The durable advantage
The cheaper report is only the first-order benefit. The second-order benefit is reuse.
If AI produces a structured evidence profile rather than a disposable narrative, the next employer, investor, bank, supplier, or counterparty does not need to restart the same work. The profile can refresh, collect context, preserve audit trail, and carry verified evidence forward.
That is how the cost curve breaks: lower synthesis cost, lower collection cost, and lower duplication.