PE manager due diligence is too high-stakes and judgment-heavy for AI to own.

There is a repeatable framework — reviewing PPMs, track records, team bios, audited financials, reference checks — but each manager presents unique strategy nuances, edge cases, and red flags that require fresh judgment. The structure is similar; the substance is not.

Success criteria are deeply ambiguous: 'is this manager worth investing with?' has no crisp answer, and reasonable experts disagree. There is no objective threshold an agent can check to know when the work is done or correct.

Critical inputs — audited financials, LP references, side letters, fund legal docs, proprietary databases like Preqin or PitchBook — are gated behind NDAs, relationships, and paid subscriptions. Reference calls and management meetings are inaccessible to an agent entirely.

A missed red flag or misread track record could lead to committing tens or hundreds of millions of dollars to a fraudulent or underperforming manager. Capital commitments in PE are illiquid and locked up for 10+ years — essentially irreversible.

Assessing whether a GP team will hold together through a downturn, whether their edge is real or narrative, and whether references are genuinely enthusiastic or diplomatically lukewarm requires human intuition, relationship context, and pattern recognition built over years in the industry.