Family Office Due Diligence & Staff Screening with AI

Two Different Due Diligence Problems, Often Confused

"Due diligence" at a family office usually means investment due diligence — strategy, track record, fees, operational infrastructure. That's real and important, and it's covered in the main Family Office AI guide. This guide covers a distinct, less-discussed problem: people due diligence — screening the humans a family office trusts with capital and access, whether they're an outside fund manager's key person or an internal hire who will have signing authority. Specialist recruitment and investigative firms exist specifically for this niche (family office staffing agencies, key-person background investigators), which is itself a signal that it's a distinct discipline, not a subset of investment analysis.

Key-Person Risk on Outside Managers

Before committing capital to a fund, LP due diligence reviews who the fund actually depends on: the named key person in fund documents, what the key-person clause actually triggers if that person departs or is incapacitated, team turnover history, and whether the current team has a track record of working together or is a recently assembled group riding a prior firm's returns. A strategy with excellent historical numbers backed by a team that has never worked together before is a materially different risk than the same numbers from a stable team — the numbers alone don't show that.

Internal Staff Screening: Pre-Hire and Periodic Re-Vetting

The higher-frequency, lower-glamour version of this problem is internal: anyone who will touch the money — investment staff with trading or wire authority, controllers, bookkeepers — and anyone with standing physical access — household staff, drivers, property managers — represents real risk regardless of how well the office knows them personally. Industry guidance on this is specific: independent pre-hire screening for anyone touching the money, and just as important, periodic re-vetting of long-tenured staff, since a clean background check at hire says nothing about a person's situation five years later. The principle behind this: genuine oversight requires someone other than the people who would be checking their own work.

People Due Diligence Components

A structured people due diligence process for either a fund manager's key personnel or an internal hire typically covers the same core components: background and criminal history screening, OSINT-based reputational and adverse-media review, sanctions and PEP (politically exposed person) list screening, and — for fund managers specifically — ultimate beneficial owner (UBO) verification to confirm who actually controls the entity being invested with. None of these are things Claude can perform directly; all of them produce documents, reports, and data points that need to be synthesized into a coherent picture a principal can act on quickly — which is where Claude's role actually sits.

The AI-Vendor-Liability Problem

A 2026-specific wrinkle worth knowing about directly: as background-screening providers increasingly build AI into their own products (automated adverse-media scoring, AI-assisted resume-to-record matching), regulatory guidance to employers has gotten explicit that the employer remains responsible for that vendor's algorithm — if a screening vendor's AI is biased or inaccurate, the family office that used it faces the consequences, not just the vendor. This means selecting a background-screening vendor is itself a due-diligence decision now, not a commodity purchase — worth documenting why a given vendor was selected, same as any other material vendor relationship.

Illustrative Scenarios

The two scenarios below are composite, illustrative examples for demonstrating the workflow — not real family offices or real individuals.

• "Draft a due diligence comparison memo for three candidate portfolio managers for a $400M single-family office CIO role. For each candidate, I'll provide: 10-year track record summary, prior firm and reason for departure, and reference call notes. Structure the memo with: (1) a side-by-side track-record comparison table; (2) a consistency assessment — does the stated track record align with what references described, and are there any gaps or discrepancies; (3) 3-5 specific follow-up questions for the next interview round per candidate, focused on anything ambiguous or unconfirmed; (4) explicitly flag anywhere the provided information is insufficient to reach a conclusion, rather than filling the gap with assumption."
• "Compare this year's periodic re-screening report for a long-tenured staff member against their original pre-hire screening report from [year]. Summarize: (1) what changed (new addresses, new outside affiliations, any new public records); (2) what stayed the same; (3) whether any change, on its face, warrants a follow-up conversation versus being routine (e.g., a home purchase is not inherently a flag; an undisclosed new business interest touching client relationships likely is). Do not speculate about intent — describe only what the two reports show and where they differ."
• "Review this fund's key-person clause and team bios for an LP due diligence file. Identify: (1) who is named as key person(s) and what specifically triggers the clause (departure, incapacity, reduced time commitment — the clause language varies); (2) team turnover over the stated track record period — how many of the people who generated the historical returns are still at the firm; (3) whether the current team has a demonstrated history working together prior to this fund, or is largely newly assembled; (4) a one-paragraph summary of key-person risk suitable for an investment committee memo."

Where to Start

For fund-level (not people-level) due diligence — strategy, fees, operational infrastructure — see the main Family Office AI guide's Alternative Investment Due Diligence section. For the broader private equity and venture diligence workflow this connects to, see Private Equity AI and Financial Due Diligence AI. For AML/sanctions screening depth, see KYC & AML Screening AI. Start with whichever of the two scenarios above is closer to a live task — the memo-comparison pattern (structure what's already been gathered, flag gaps, don't fill them with assumption) is the reusable core across both.

Frequently Asked Questions

Can Claude run a background check on a family office job candidate?

No. Claude has no access to background-check databases, criminal records, credit bureaus, or sanctions lists, and cannot itself screen anyone. Its role is downstream of that: synthesizing information a licensed screening provider or investigator has already gathered into a structured due-diligence memo, flagging inconsistencies across documents, and drafting the write-up a principal reviews before a hiring or investment decision.

Why do family offices need to screen their own staff, not just outside fund managers?

Industry due-diligence practice notes that the largest losses at family offices come from people closest to the assets, not strangers — investment staff, controllers, and household staff who have standing access to accounts, wire authority, or physical property. Pre-hire screening plus periodic re-vetting of long-tenured staff closes a gap that one-time hiring checks miss, since risk (financial pressure, external relationships, access creep) can change years into a tenure.

What is key-person risk in a fund due diligence review?

Key-person risk is the exposure a fund carries if a named critical individual (often the lead portfolio manager or founding GP) departs, becomes incapacitated, or is otherwise unable to continue. LP due diligence typically reviews the fund's key-person clause, succession provisions, team turnover history, and whether team members have a track record working together, since a fund's stated strategy is only as durable as the people executing it.

Are family offices legally liable for a background-check vendor's AI?

Increasingly, yes in substance if not always in a single bright-line rule — 2026 guidance to HR and compliance teams is explicit that a business remains responsible for a screening vendor's algorithm, including bias or accuracy problems in an AI-assisted background check product. A family office selecting a screening vendor should treat that vendor selection itself as a due-diligence decision, not just approve trust because the vendor uses AI.