The AI Position: What Buyers Expect from PE-Backed Portcos in 2026

Two years ago, buyers of PE-backed businesses didn't ask about AI in due diligence. Today they do, and the question has already evolved. It used to be "what's your AI strategy?". Now it's closer to "what's your AI position, and what value is it creating?". That shift matters. It reveals how buyers are reading AI adoption. It's not primarily as a technology question.

The shift buyers have made

Sit through a technology DD process today and you'll notice something specific about how buyers approach AI. They rarely ask about the technology itself. They ask about the pattern of adoption — what's been prioritised, who owns it, what's been delivered, what's been deliberately not done.

That pattern of questioning tells you what's really going on. Buyers have started reading AI adoption as a proxy for something else: how the business makes decisions.

The claim isn't that AI adoption doesn't matter on its own terms. Specific use cases with quantified commercial value are valuable, and buyers will note them. But increasingly, buyers care as much about how the business got there as they do about the tools deployed. Because how a business adopted AI reveals how that business makes decisions across every function.

Why AI is a useful signal specifically

Every business decision reveals something about management. But AI decisions reveal more than most, for three reasons.

AI is a novel decision context. It's recent enough that leadership teams can't fall back on inherited playbooks. Every business is deciding for itself — which tools to buy, which use cases to prioritise, what policy to write, who owns it. What they decide reveals how they decide.

AI is cross-functional. Adoption touches product, operations, finance, sales, marketing, HR, legal, IT, and security. A leadership team that makes coherent AI decisions across those functions is demonstrating something rare: genuine cross-functional coordination. A team that makes fragmented AI decisions is demonstrating the opposite.

AI is information-rich in a small sample. Ten minutes of conversation about AI reveals more about management maturity than an hour on most other topics, because the "right" answer isn't obvious yet, so the business has to reveal its actual reasoning.

For a buyer with two or three weeks of DD access, signals like that are gold. They can't observe every function directly. They rely on shortcuts. AI has become one of the sharper ones.

What AI adoption actually reveals

Five specific things buyers can infer from looking at a portco's AI adoption pattern.

1. Decision-making discipline

Weaker signal: Every department bought its own AI tool. No coordination between them. Total AI spend across the business is invisible or fragmented across departmental budgets.

Stronger signal: AI purchases went through a coordinated review process. Spend is visible. There's a rationale for what's centralised and what's decentralised.

What this tells the buyer: the business has functioning governance. For AI, and by extension for everything.

2. Prioritisation quality

Weaker signal: "We're exploring lots of AI use cases across the business." No ranking, no chosen bets.

Stronger signal: "We've identified three high-value use cases, deployed one, learned from it, and here's what we're going to do next."

What this tells the buyer: leadership can choose. And choosing means saying no to things, which is genuinely rare and hard.

3. Speed of learning

Weaker signal: "We've been evaluating AI for twelve months and are still in discovery."

Stronger signal: "We ran three pilots in six months. Two got killed for good reasons. One is in production and here are the outcomes."

What this tells the buyer: the organisation can move. It can also kill things, which is arguably more important than being able to start them.

4. Risk-opportunity balance

Weaker signal: Either paranoid, having banned all AI tools with no substitute path. Or reckless, with no policy, allowing anyone to use anything, no oversight of data going into third-party models.

Stronger signal: A documented AI Usage Policy that permits specific use cases with clear guardrails, applied in practice, understood by teams.

What this tells the buyer: management can hold two things in tension without collapsing to one extreme. Not risk-averse, not risk-blind. Grown-up.

5. Ownership clarity

Weaker signal: "AI is everyone's responsibility." Or: "We're going to hire a Head of AI."

Stronger signal: "Our COO owns AI adoption for operational functions. Our CTO owns technical governance. Our General Counsel owns policy. Here's how they coordinate."

What this tells the buyer: the business understands accountability. Real ownership, not slogans.

The pattern in practice: a recent portco example

In a recent PE-backed engagement, I was brought in as interim Technology Director for a professional services firm with US and EMEA operations, preparing for exit within a few years.

When I started, AI was already in use across the business. Various team members had signed up to ChatGPT, Claude, and specialised marketing and coaching tools. Some of it was clearly value-adding — content drafting, research support, meeting summarisation. Some of it was creating risk quietly in the background. Client data was going into third-party tools with unclear terms, and no policy on what could and couldn't be shared.

None of this made the leadership team careless. What it made them was uninvolved. AI hadn't yet been recognised as something requiring management attention. There was no policy, no governance, no coordination, no visibility of spend. Adoption was happening, but nothing revealed how the business was choosing to adopt.

From a buyer's perspective in DD, that state would have been informative. Not because the tools themselves were problematic. Most of them were fine. It was because the absence of any decision-making about them was itself a signal. A buyer would have read it as: the leadership team is not yet operating cross-functionally on emerging technology. And that inference would have applied more broadly than just to AI.

The work over the following weeks was straightforward. Draft the org's first AI Usage Policy. Identify the two or three highest-value use cases and put owners on them. Introduce visibility of AI spend. Bring the pattern of adoption under management attention.

None of that was technically complex. But it changed the signal the business would send. Not by inventing capability, but by naming what was already there, the good and the risky, and putting a management wrapper around it.

What a credible AI position looks like

Pulling this together, what should a portco actually have in place to give buyers a credible AI position? Five components.

A documented AI Usage Policy, approved by leadership, understood by teams, applied in practice. It doesn't need to be long. A short policy that's actually followed is stronger than a comprehensive one that isn't.

Named ownership. Someone at leadership level is accountable for AI adoption. Not "everyone" and not "the future Head of AI." A named individual or small group, with defined scope.

A prioritised use case list of three to five specific use cases identified as high-value, with commercial rationale. Some deployed, some in pilot, some deliberately not being pursued (with reasons).

Delivered outcomes with numbers. At least one AI-enabled improvement live in the business with a measurable outcome attached. Efficiency gain, capacity release, revenue impact, error reduction, something quantified.

Visibility of AI spend and data flows. The business knows what it's spending on AI in aggregate and what data is going into which tools. This is the DD-relevant governance layer.

None of this is complex. It's not expensive. But few PE portcos have all five in place. That's precisely why the businesses that do stand out.

Where a strategy still comes in

Everything above might sound like it's arguing that AI strategy doesn't matter, that only management execution does. That would be wrong.

A strategy still requires decisions. Which use cases to prioritise. What to build vs buy. Which tools to standardise on. How to invest in data readiness. What to explicitly deprioritise. Those are strategy questions and they need answers.

The distinction is what the answers look like. A defensible AI strategy in 2026 is not a forty-seven-slide deck describing a future state. It's a short, specific document, sometimes just a page or two, that names the current position, the immediate priorities, the named owners, and the boundaries of what's not being pursued. It reads as a working artefact, not a marketing piece.

Buyers can tell the difference within one conversation.

What buyers actually ask

To make this concrete, here are the specific AI questions I've seen buyers ask in recent portco DD processes:

• What is your AI Usage Policy? Who approved it? When was it last reviewed?
• What AI tools are in use across the business today, and by whom? What is the total spend?
• Which AI use cases have been deployed with measurable outcomes? What are those outcomes?
• What data is going into third-party AI tools, and how is that governed?
• Who owns AI adoption in the business? What is their scope?
• What AI use cases have you deliberately chosen not to pursue, and why?

Notice the pattern in those questions. Only the first is about the policy artefact itself. The rest are about the pattern of decision-making the business is displaying. That is not accidental.

Where to start

If you're a portco CEO, CFO, or an Operating Partner reading this and wondering where your business stands, the fastest way to get an honest view is the same structured self-assessment used for all six pillars of technology DD readiness. Pillar 6 of the DD Readiness Scorecard covers AI specifically: six questions, scored 1 to 5, twenty minutes to complete honestly.

The scorecard is freely shareable, no lead capture required. Score your business honestly and see where the gaps sit. Score three or more of the six AI questions at 2 or below, and there is work to do.

The through-line

The buyer's question about AI has changed because buyers have realised something specific: AI adoption is a management-capability question first, a technology question second. How a business is adopting AI reveals how that business makes decisions across every function: the discipline, the prioritisation, the ownership, the learning speed, the risk-opportunity balance.

The good news is that a credible AI position is not expensive to build. Policy, ownership, prioritised use cases, delivered outcomes, spend visibility. None of these require large investment. What they require is management attention. Where a leadership team gives AI that attention, the resulting artefacts speak for themselves.

Where they don't, the absence itself becomes the signal.

James Scott is the founder of The Clarity Partnership, an interim and fractional practice for PE-backed and privately-owned businesses at inflection points. Nearly thirty years of transformation, integration, and enterprise programme delivery across Unilever, Gravity Media, Samsung, General Mills, and PE-backed portfolio companies.