AI in Private Equity: What Top Firms Are Actually Doing

AI in private equity is being used in two places at once: inside the firm, to speed up diligence, modelling and memo-writing, and inside portfolio companies, where it does the heavier lifting of actual value creation. The firms getting real returns treat the second as the prize and the first as the warm-up. This guide covers what AI in private equity looks like in practice, where it pays off, and why the winners aren't following a playbook.

This is for GPs, operating partners and portfolio leaders who keep hearing that AI will reshape private equity and want the concrete version, not the conference-stage version. We train investment teams and their portfolio companies on these tools, so the examples here are what we actually see, including a few you'll recognise.

Start with the thing that makes private equity different from every other buyer of AI: the motivation.

What private equity actually wants from AI

A marketing team wants AI to save time. A private equity firm wants something narrower and larger. It doesn't care about AI for its own sake. It cares about EBITDA, multiples and exits, and it has worked out that AI is one of the biggest available levers on all three.

"PE understands AI is a huge opportunity to drive up company value. They want to sell at a higher multiple, they want a bigger exit, and rolling out AI operations across their firm and their portfolio is a surefire way to achieve that."

— Alex Kantjas, 9x co-founder

That reframes everything. The question a firm is really asking isn't "how do we use ChatGPT?" It's "where does AI move enterprise value across a portfolio, and how fast can we get there?" Everything below ladders up to that.

The two places AI shows up

AI in private equity lives at two levels, and conflating them is the most common mistake.

The first is the firm itself: the investment professionals, the deal team, the operating partners. Here AI speeds up the work of finding, evaluating and managing deals.

The second is the portfolio: the operating companies the firm owns. Here AI changes how the business actually runs, and this is where the value that shows up at exit gets created.

Both matter. But they are not equal, and the best firms are clear about which one moves the number.

At the deal desk: faster diligence and modelling

Inside the firm, the wins are immediate and easy to feel, which is why most firms start here. AI reads data rooms, drafts investment memos, summarises management calls, builds first-pass models and pressure-tests assumptions. Work that used to eat an analyst's week compresses into an afternoon.

"The most convincing thing? When they see Claude just did an LBO forecast in 30 minutes that took their analyst a week."

— Alex Kantjas, 9x co-founder

Take that seriously and the implication is uncomfortable in a good way. If a first-pass leveraged buyout model is a 30-minute job, your team's scarce hours move off the mechanics and onto judgement: which deals to chase, which assumptions to challenge, what the model can't see. The tools that do this well are Claude for the analysis and document work, and Claude Code when a repeatable model or data pipeline is worth building once and reusing across deals.

The caution here is real. A machine-built model is a draft, not a decision. Someone senior still owns the numbers, the diligence and the recommendation. AI shortens the distance to a strong first version; it doesn't remove the person accountable for it.

The larger fund-level prize isn't speed, it's reach. A fund that screens a couple of thousand companies a year and dismisses most of them can point AI at the top of the funnel, seeing many times more targets and choosing from a far wider field. When the binding constraint is capital rather than deal flow, more good options at the same cost is a direct edge on returns.

In the portfolio: where the value is actually made

Speeding up the deal desk is worth doing. It is not what moves the multiple. The value that shows up at exit is created inside the operating companies, and the firms pulling ahead know it. This is no longer fringe: Bain's survey of investors managing $3.2 trillion found nearly a fifth of portfolio companies have put generative AI into real use with concrete results.

The difference is stark when you name it. A partner processing board decks a little faster is a rounding error. A portfolio company whose teams build their own tools, automate their own processes and ship without waiting on a backlog is a different business by the time you sell it. That is the prize, and it's a portfolio-wide capability, not a headquarters convenience. The data backs the instinct: in McKinsey's analysis of PE-backed companies, the most AI-mature traded at the highest revenue multiples, a median of 31 times revenue.

The clearest picture of where this goes is a company already living it. Ours is a subscription business years ahead of the field: business teams building their own tools, an open budget to spend on AI usage, and process owners automating their own work rather than queuing for engineering. Non-engineers who have never written code ship internal tools and automations after a couple of years of working this way. That is what an AI-native operating company looks like, and it's the model a smart sponsor pushes across a portfolio.

What the aggressive firms are doing

The leading edge isn't subtle about it. Some of the largest firms have moved AI from experiment to infrastructure.

That's the tell. When a firm rolls out secure, enterprise-grade AI across its people rather than letting a few analysts dabble, it has stopped treating this as a tool and started treating it as a capability. And there's a social dimension the numbers miss: in a market this competitive, the firm that has trained its whole team and portfolio can say so, and the one that hasn't started feels it. Demonstrable AI proficiency is quietly becoming table stakes for how sophisticated a sponsor looks to its own LPs and management teams.

It isn't only the mega-funds. RM Equity Partners, a European fund built on a strict rule of no centralised, cross-portfolio initiatives, broke that rule for AI: it ran a single enablement programme across 15 portfolio companies at once, trained more than 600 operators, and finished with a three-day hackathon that shipped over 25 automations into production. One of them saved a single portfolio company around €40,000 a year.

"AI is to knowledge workers what the steam engine was to manufacturing."

— Eugen B. Russ, Managing Partner, RM Equity Partners

That is what treating AI as infrastructure rather than a pilot looks like at portfolio scale.

The honest part: there is no playbook

Here is where most AI-in-private-equity content oversells, and where the credible version earns trust instead. There is no neat, repeatable transformation playbook you can buy and roll out. Anyone selling one is selling comfort.

"There's this expectation right now that there's a playbook to transform companies with AI... There's no playbook. Nobody knows what they're doing. You have to figure out your own unique playbook—that work has to be done."

— Alex Kantjas, 9x co-founder

That isn't a reason to wait. It's a reason to start, because the firms that get good at this are learning by doing it, on real deals and inside real portfolio companies, while their competitors wait for a method that isn't coming. The realistic expectation is a year of building, adjusting and fixing, not a switch you flip. The advantage compounds for whoever begins.

How to start without overcommitting

You don't need a firm-wide mandate to begin. The sensible path has three moves.

First, train the investment team on the real work: diligence, memos, modelling. It builds fluency fast and it's low-risk, because a person still signs off on everything.

Second, and more important, pick one portfolio company that's willing and point the same capability at its operations. That's where you'll see value that actually reaches the exit.

Third, treat it as a capability you're building, not a project you're finishing. Save what works as repeatable agentic workflows, train people rather than tools, and expand from the wins. If this is landing on one person across the firm, you're describing a role worth naming, the Chief Automation Officer.

The bottom line

AI in private equity works on two levels: it speeds up diligence and modelling inside the firm, and it creates real value inside portfolio companies. The first is easy and worth doing; the second is where the multiple moves. The aggressive firms have made AI infrastructure rather than experiment, they've accepted there's no playbook, and they're compounding an advantage by starting now. Point it at the firm to build fluency, point it at a portfolio company to build value, and treat the whole thing as a capability rather than a one-off.

If you want your firm or your portfolio companies building this properly, our hands-on Claude training for teams takes investment professionals and operators from their first task to workflows that run the work.

Common questions

How is AI used in private equity? At two levels. Inside the firm it speeds up diligence, financial modelling, memo-writing and management-call analysis. Inside portfolio companies it automates operations and lets teams build their own tools, which is where value that reaches exit is created.

Does AI actually create value in a portfolio, or just save time at the firm? Both, but they're different sizes. Time saved at the deal desk is useful; value created inside operating companies is what moves the multiple. The firms getting real returns prioritise the portfolio.

Which AI tools are private equity firms using? General-purpose tools like Claude and Cowork for analysis, documents and automation, and Claude Code where a repeatable model or pipeline is worth building. The largest firms are rolling out secure, enterprise versions across their teams.

Is it too early to invest in AI for private equity? No. There's no finished playbook, but that favours early movers: the firms getting good are learning by doing on real deals and portfolio companies while others wait.