Executive briefing: AI in transactions
AI Due Diligence for M&A: What Buyers and Sellers Must Check
AI due diligence tests a target's AI for risk, readiness and value before the deal closes. It asks what AI the business really runs, whether it holds the rights to its data and models, what regulation applies, and whether the operating model can deliver the value in the deal case. In AI due diligence for M&A, each finding maps to a deal lever: the price, a specific warranty or indemnity, a condition before completion, or a funded fix in the 100-day plan.
Read the full briefingIn brief
- AI due diligence looks through three lenses: risk, readiness and value. Most diligence stops at risk.
- Eight areas can move the price, the warranties or the deal structure, from the AI inventory and data rights to key people and overstated AI claims.
- Ask for evidence, not descriptions: contracts, logs and numbers, not slides and adjectives.
- Red flags do not always kill a deal. They should change the price, the protections or the plan.
- Sellers who can evidence their AI protect their price. Every claim they cannot evidence becomes a discount or a warranty.
Why the board should care
AI claims are now part of the price
An information memorandum that calls a business "AI-powered" is asking the buyer to pay for it. If the AI runs in a demo rather than in production, the buyer is paying for a slide.
Rights problems travel with the deal
Data of unclear origin, unclear ownership of models and outputs, and reliance on one provider with no exit terms all pass to the buyer at completion.
The value is delivered after completion
The deal case depends on the operating model the buyer inherits. If nobody owns the 100-day plan, the value in the model stays in the model.
Full briefing
Read the full briefing
Free for executives. Confirm your work email once and every AIPEX executive briefing opens for 12 months.
- The video briefing
- The full written briefing
Next step
Find the gaps before a buyer prices them in
The AI Readiness Assessment: Full Report (£999 + VAT) looks at 30 areas across six domains, in your own business before a sale or in one you have just acquired, with a 60-minute debrief with a senior partner. Prefer to talk first? A free 30-minute scoping call with a senior partner.
FAQ
Frequently asked questions
What is AI due diligence?
AI due diligence tests a business's AI for risk, readiness and value before a transaction closes: what AI it really runs, whether it holds the rights to its data and models, what regulation applies, and whether the operating model can deliver the value in the deal case.
How is AI due diligence in M&A different from technology due diligence?
Technology diligence covers some of the risk, through IP, data protection and systems. AI due diligence also tests readiness, whether the AI runs the business or sits in pilots, and value, whether the operating model can deliver the upside in the deal case.
What AI risks should a buyer check in a target?
Eight areas: the AI inventory including shadow AI, data rights, IP ownership of models and outputs, dependence on one model provider, regulation such as the EU AI Act and GDPR, security and incident history, key people, and whether the AI claims in the information memorandum are real.
Which AI red flags should change the price of a deal?
AI claims with no production evidence, training data of unclear origin, one provider with no exit terms, capability held by two or three people, no inventory, owner or policy, and value that is claimed but never measured. Each should lead to a price adjustment, a warranty or indemnity, a condition before completion or a funded fix in the 100-day plan.
How should a seller prepare AI for due diligence?
In four steps, before the information memorandum is written: inventory every AI system and its owner, put the evidence into the data room, measure value against a baseline, and fix gaps such as shadow AI, key-person risk and missing controls.
What is a 100-day AI integration plan?
Three sprints after completion. Days 1 to 30 stabilise and protect, days 31 to 60 capture quick value, and days 61 to 100 set up a common AI operating model, a shared data and platform plan and a 12-month plan for the board.
Who should own AI after an acquisition?
A named Chief AI Officer should own the 100-day AI integration plan and report to the board on the value in the deal case. For many businesses a fractional Chief AI Officer is the right fit.
This briefing is general information on AI in mergers and acquisitions. It is governance guidance, not legal advice. AIPEX is not a certification body. It does not replace legal, financial or technical due diligence; take advice from qualified advisers.