AI DUE DILIGENCE

From data room to decision-grade review.

Kapintelligence AI due diligence turns a fund data room into a complete, source-linked review: DDQ completion, financial model extraction and recalculation, red-flag identification and an IC-ready decision record. It covers funds, co-investments and single-asset opportunities. Analysts validate every finding before it is final.

Assess funds, co-investments and single asset opportunities. Use standard DDQs and apply your own investment criteria, then build the {{ recalcWord }} and IC memo - with analysts and managers validating before anything is final.

Book a Demo Explore a Sample IC Pack
app.kapintelligence.com - fund DDQ run
Kapintelligence fund DDQ run: a question with its answer, source attachment and sign-off state
Kapintelligence fund DDQ run: section completion by area, question cards with answers, attachments and sign-off states

A live fund DDQ run. Section completion on the left, answered questions with their attachments and sign-off state on the right.

First reviewable output in hours

Coverage, critical knock-out questions and first source-cited answers, subject to data-room scope.

Full institutional review: typically under 30 days

End-to-end process range, not software processing time. It depends on data-room completeness, manager response times and your own review steps.

Nothing is final without a person

Every draft carries a validation state: draft, manager-confirmed, analyst-approved or exception.

REVIEW DEPTHS

From quick screening to full due diligence.

{{ m.name }}

{{ m.scope }}

Promise
{{ m.promise }}
THE WORKFLOW

Six steps, each with a visible state.

Select a step to see what the workspace produces at that point - and what your team gets out of it.

{{ stepProgressLabel }}
Step {{ step.num }}

{{ step.label }}

{{ step.desc }}

{{ step.panel }}
{{ r.l }} {{ r.v }} {{ r.t }}
COVERAGE SIMULATOR

See what a data room can and cannot answer - before you start reading.

Add or remove document types and watch which DDQ and IC requirements lose their evidence. This is the first thing the workspace does with a new data room.

Documents in the data room - select to add or remove
{{ score }}
{{ scoreLabel }}
Coverage of 43 DDQ and IC requirements
{{ r.label }} {{ r.status }}
Generated manager request
{{ gapCount }}
{{ g.ask }}
{{ g.why }}
Every requirement has a source

Coverage is complete for this requirement set. Analysis moves to contradictions between documents and to the calculations that need verifying.

The workspace drafts the request list; the analyst decides what to send. Illustrative requirement set based on a European buyout DDQ.

FINANCIAL VERIFICATION

Reported figures, recalculated figures, and the difference between them.

{{ verifyIntro }}

Sample Buyout Fund II · vintage 2018 · as at 31 March 2026 · illustrative
{{ p.label }}
{{ p.value }}
{{ p.delta }}
{{ reconTitle }}
{{ r.item }} {{ r.effect }}

{{ methodNote }}

{{ step.note }}

What you get
{{ step.out }}

Illustrative - anonymised fund, synthetic figures.

UNGATED

Read a full IC pack before you talk to us.

Anonymised and illustrative, with every important statement showing its validation status and evidence link - the same structure your committee would receive.

Open the sample pack
{{ p }}
GOVERNED, NOT JUST GENERATED

A writing assistant can draft. An institutional workflow must govern.

General-purpose models are useful, and analysts already use them. The difference is what a committee and an auditor can rely on afterwards.

Dimension General-purpose AI assistant Kapintelligence workflow
{{ c.dim }}
General-purpose AI assistant {{ c.generic }}
Kapintelligence workflow {{ c.kap }}

A comparison of workflow properties, not of model quality. Kapintelligence uses large language models under contractual no-training terms; the governance around them is the product.

See Kapintelligence in action.

From fund due diligence to IC ready output.

Book a Demo Explore a Sample IC Pack
app.kapintelligence.com - fund DDQ run
Kapintelligence fund DDQ run, full view
Pinch to zoom · drag to pan

Related: See how the platform connects due diligence to portfolio reporting, learn how institutional investors use Kapintelligence for fund selection, or review security, data residency and access controls. Fund managers can also use the platform for their own deal assessment.

How does Kapintelligence turn a data room into a decision-grade review?

Assess funds, co-investments and single asset opportunities. Use standard DDQs and apply your own investment criteria, then build the recalculated model and IC memo - with analysts and managers validating before anything is final.

Timing

Review depths: From quick screening to full due diligence.

The workflow: Six steps, each with a visible state.

  1. Assess coverage – Classify the data room against required documents, DDQ sections and IC evidence; identify missing or conflicting material. Output: Coverage map and a drafted manager request list.
  2. Extract and structure – Populate questions, figures, clauses and entities with their source passages attached. Output: A populated DDQ with a source reference per answer.
  3. Analyse and recalculate – Apply methodology, policy checks, knock-out criteria and red-flag rules; calculate supported metrics. Output: A reconciliation to GP-reported figures, with the variance explained.
  4. Interrogate – Ask questions across the data room; generate manager follow-ups from gaps and contradictions. Output: One consolidated manager request list, ready to send.
  5. Validate – Track manager confirmation, analyst review, corrections and approvals. Output: An auditable approval chain per answer and per figure.
  6. Prepare the decision record – Generate the completed DDQ, model and IC memo with open issues and source lineage preserved. Output: An IC-ready decision record in your own memo format.

What can a coverage simulator reveal about your data room?

Add or remove document types and watch which DDQ and IC requirements lose their evidence. This is the first thing the workspace does with a new data room.

Document types: Manager DDQ response, Limited Partnership Agreement, Audited financial statements, Quarterly reports, Track-record schedule, Cost information sheet, Risk policy and compliance manual, Firm structure chart.

How does financial verification compare reported and recalculated figures?

Where the source documents support it, key fund metrics are recalculated from the underlying cash flows with a visible methodology and reconciled against the figures the manager reported.

Sample Buyout Fund II, vintage 2018: GP-reported Net IRR 18.4% vs recalculated 16.9% (-1.5pp). GP-reported TVPI 1.92x vs recalculated 1.88x (-0.04x). DPI 0.71x reconciles. RVPI reported 1.21x vs recalculated 1.17x (-0.04x).

Read a full IC pack before you talk to us.

Anonymised and illustrative, with every important statement showing its validation status and evidence link. Contents: Executive summary, Investment thesis, Fund and manager facts, Track-record analysis, Portfolio construction, Terms and economics, Key risks, Operational and legal findings, Open questions, Recommendation inputs, Appendices and sources.

How is AI output governed and validated?

A writing assistant can draft. An institutional workflow must govern. General-purpose models are useful, and analysts already use them. The difference is what a committee and an auditor can rely on afterwards.