Private Equity and AI Value Creation

AI has become a fixture of investment theses, and for good reason: it offers real operational leverage in the right businesses. It has also become a source of narrative that does not always survive contact with reality. For operating partners, the challenge is to distinguish genuine, deliverable value creation from the appearance of it — both when assessing a target and when driving value across the hold period.

The discipline is the same one private equity applies everywhere: separate what is claimed from what is evidenced, and understand what it will actually take to realise the upside. AI does not change these questions; it simply adds a domain in which they are easy to answer loosely.

In diligence: capability, not claims

When a target presents AI as a source of value or differentiation, the useful questions are concrete. Is the capability real and in production, or a pilot dressed as a platform? Does it depend on data the business actually owns and can use? Is it defensible, or trivially replicable by a competitor with a subscription? A clear-eyed assessment of technical maturity and governance often reveals a gap between the story and the substance.

• Is the AI capability in genuine production use, and what would it cost to sustain and scale?
• How dependent is it on specific individuals, suppliers or data that may not transfer?
• What governance and risk controls surround it, and would they satisfy an acquirer or regulator?
• Is the claimed advantage defensible, or available to any competitor off the shelf?

Separate what is claimed from what is evidenced, and understand what it will actually take to realise the upside.

Across the hold: value that compounds

The more durable opportunity is often not in the deal thesis but in the operational improvement AI can drive across a portfolio company during ownership — in efficiency, in the productivity of knowledge work, and in the quality of decision-making. Realising it requires the same rigour as any value-creation plan: a small number of prioritised initiatives, clear ownership, and honest measurement against a baseline.

Govern the downside

AI also introduces risks that can erode value quietly: mishandled data, unmanaged model dependencies, and adoption that outruns control. An operating partner protecting an investment should expect the same governance discipline around AI as around any other material risk — not to suppress ambition, but to ensure that value created is not offset by exposure created alongside it.

Independent assessment pays for itself

Because the market is loud with AI narrative, an independent, technically grounded assessment is disproportionately valuable in this domain. Asking the hard questions early — in diligence and again as value-creation plans are set — is a modest cost against the value that a misjudged AI thesis can quietly consume.