Who leads AI at private equity firms

The AI lead is the newest seat in private equity. Of the leads whose start date we could find, seven in ten took the job in 2025 or 2026, and more than half started in 2026 alone.

In our first post, we counted which firms have an AI lead. The answer depended on size: four in ten firms managing more than $10 billion have one, and in the lower mid-market it’s one in twenty. That post ended on a question: how does a smaller firm get the capability without the headcount? To answer that, we first needed to know what the headcount looks like.

So this time we looked at the people: what they’re called, where they came from and what they work on. The short version is that the job is new everywhere, and it shrinks in seniority and scope as firms get smaller.

A brand-new job

We found a public start date for about half of the AI leads. Of those, 53% started in 2026 and another 16% in 2025. Only about one in five has been in the seat since before 2024.

What’s new is AI itself. Among leads with AI in their title, 81% started in 2025 or 2026, and 58% started this year alone. The year isn’t over: at its pace through September, 2026 would end with more than three times as many new AI-titled leads as 2025. The rest are heads of data science, digital and technology who now carry the AI remit, and their roles are older: 60% started before 2025, against 19% of the AI-titled leads. Many of those jobs existed before generative AI and have since absorbed it.

The surge shows up at every size of firm. Of all dated leads, 64% at mega-funds started in the last two years, 80% at large firms, 67% at mid-sized firms and more than half in the lower mid-market.

Private equity is part of a wider wave. IBM’s 2026 CEO study, which surveyed 2,000 CEOs, found 76% of organizations have a chief AI officer, up from 26% in 2025. Recruiters say private equity has moved early. In early 2025, Korn Ferry’s global head of data and AI said PE was taking the lead on AI recruitment by creating AI operating executive roles that assess portfolio companies for where AI can improve processes. The hires keep coming. In August, Great Hill Partners named Sam Liu its Director of AI, hired from an Associate Partner role in Bain & Company’s AI practice.

So almost nobody in this job has taken an AI program all the way through a portfolio yet. Most are still in their first year of working out what the job is.

Who gets hired

The typical PE AI lead is a technical manager. Of the leads whose depth we could judge from public sources, 57% manage technical teams, a third are business or strategy people, and only one in ten is a hands-on ML engineer.

They arrive from everywhere. Counting every earlier employer type in each career:

The most common single path, a quarter of the careers we could classify, is the data science practitioner who moved into management. CIO/CTO-type operators and ex-consultants follow at about one in eight each. Fewer than one in ten are best described as big-tech hires, and about one in fifteen as founders.

They are well credentialed. Of the leads whose education we could find, 13% hold PhDs and 58% a master’s or MBA. Where the subject is stated, four in five studied a technical field such as computer science, engineering, maths or statistics.

Background shapes the program. In the largest survey of PE investors to date, Gompers, Kaplan and Mukharlyamov found that firms use different mixes of financial, governance and operational engineering, and that these strategies are strongly influenced by the career histories of the firm’s founders. The same likely holds for AI programs. The profile here, technical managers who have worked inside operating companies, fits a job that is mostly about changing how a business works. CEOs see it the same way: 83% of those surveyed by IBM say AI success depends more on people’s adoption than on technology.

And they are mostly quiet. 62% have no public talks, podcasts or bylined articles at all. Whatever they are learning stays inside their firms.

The job shrinks with the firm

“Head of AI” covers everything from a managing director with a team to a single analyst. Search firms see the same spread. Egon Zehnder writes that many firms are discovering that AI leadership is not a single, uniform role. Seniority tracks firm size closely:

Above $2 billion, 39% of AI leads are partners or managing directors. Below $2 billion, it’s 28%. The share who are VPs or below moves the other way, from 25% to 38%. The rest are principals or directors, about a fifth at both sizes, plus Chief AI Officer-type titles whose rank isn’t stated and a few we couldn’t place. At a mega-fund, the AI lead is most often a managing director (46%). In the lower mid-market, 44% are VPs, managers or individual contributors. Pay postings put a price on that spread: Ares advertised its Managing Director, Head of Data and AI at $375,000–425,000 base, while a VP data scientist at Blackstone was offered $175,000–200,000.

The remit narrows too. Above $2 billion, 77% of leads work on the portfolio. Below, 55% do. Part of that gap is bios that don’t say what the person works on. Counting only leads whose remit is stated, it’s 88% against 74%. Either way, smaller firms point less of their AI lead at the portfolio, which is where the hire is supposed to pay off.

Some of that pull toward the firm’s own work is rational. In the Bain and StepStone 2026 GP survey, GPs reported the highest returns from generative AI in deal sourcing and due diligence. Within portfolio companies, benefits skewed toward cost savings, and 39% of GPs did not expect material financial impact from AI in 2026.

Across all firms, 36% of leads work only on portfolio companies, 32% split their time between the portfolio and the firm’s own investing, and 14% work only inside the firm. Fewer than one in five bios don’t say. Most firms have one person: 69% list a single AI lead, 23% list two and 8% list three or more.

All of this describes firms that have an AI lead, and most don’t. Of the firms above $2 billion we researched, 25% have a dedicated AI lead: one in four. Between $100 million and $2 billion, it’s 6%, about one in sixteen. By size, that’s 39% of mega-funds, 19% of large firms, 8% of mid-sized firms and 4% of the lower mid-market.

Put all three together, having a lead, its seniority and its remit, and the drop-off is steep:

Picture two firms: one managing $1 billion in private equity, the other $5 billion.

At the $1 billion firm, there’s about a one-in-thirteen chance anyone’s job is AI. If someone’s is, fewer than one in three is a partner or managing director, and it’s a coin flip whether they work on the portfolio at all.

At the $5 billion firm, the odds of an AI lead are one in five. One in three of those leads is a partner or MD, and three in four work on the portfolio.

Across the whole market, about one in thirteen firms above $2 billion has a partner-level AI lead working on its portfolio. Between $100 million and $2 billion it’s about one in 110.

In their own words

Two years ago, AI leads were rare outside the largest firms. Today those firms have managing directors with teams and results to show, and their public statements read like a playbook.

• Start with the business problem: Jesse Thomas, Advent; Matt Katz, Blackstone; Tim Kiely, Baypine.
• Aim past efficiency: Dominic Gallello, Bridgepoint; Pritesh Patel, then at GoHealth.
• People keep the decision: Brian Goffman, TPG; Arezu Moghadam, J.P. Morgan Asset Management.
• It’s already in production: Lou D’Ambrosio, Goldman Sachs Asset Management; Arvind Battula, Quantum Capital.

Nearly every voice here comes from a large or mega-fund. And since three in five AI leads say nothing in public, these quotes are most of what anyone outside those firms can learn about how the job is done. The playbook is being written at the top of the market, and it is staying there.

Why the gap compounds

A firm that learns which AI plays work in pricing, field operations or claims takes that playbook into its next deal. Private equity has always worked this way. In a survey of more than 15,000 firms, Bloom, Sadun and Van Reenen found PE-owned firms are better managed than government, family and privately owned firms, with their strongest edge in monitoring practices such as lean manufacturing and continuous improvement.

AI is now part of that toolkit, and the stakes are rising on both ends of a deal. On the way in, Bain argues “12 is the new 5”: today’s deals demand faster EBITDA growth, and the winning firms will invest in talent and AI. On the way out, EY’s 2026 exit readiness study finds AI strategy emerging as a key exit differentiator, with buyers assessing AI adoption, disruption exposure and its role in future value creation. EY’s Europe West private equity leader, Cord Stümke, says buyers want embedded AI use cases and a proven record of how they translate into EBITDA. Grant Thornton reports the same pressure showing up at the deal table.

A firm without the playbook bids for the same companies with a thinner value-creation plan, and sells to buyers who will ask what it did about AI.

What this means

A lower mid-market firm can afford an AI hire, but not the one who has already done the work. The hire will probably be a VP, probably new to PE and probably working alone. That one person has to cover deal work, back office and a dozen portfolio companies at once.

A smaller firm can’t close the gap by copying the largest firms. It can’t hire a managing director with a team, and one VP spread across a dozen companies won’t get far.

There’s still time. Seven in ten AI leads have been in the job less than two years. Nearly 40% of GPs don’t expect AI to move portfolio financials this year. Nobody has a finished playbook yet, and the largest firms’ head start is a year or two at most.

What a smaller firm needs is the experience, applied inside each company for the weeks it takes to land a win. That experience can come from outside the payroll, and the evidence favors it. In MIT’s 2025 study of enterprise AI deployments, tools built with external partners reached deployment about 67% of the time, against about 33% for tools built in-house: twice the success rate. Hiring someone who has already done this work is a $350,000–450,000 annual commitment before anyone has looked at a single company. An assessment is a small fraction of that, applied to one company, and credited against the work that follows.

That’s what Talas does. We assess each portfolio company to find where AI will move EBITDA, then put engineers inside the business to build it alongside the people who do the work. The firm gets the result of an AI lead and a team, sized to a lower mid-market budget.

How we did this

We started from the PE sponsors in our first post’s dataset, drawn from SEC Form ADV filings, whose public materials show a dedicated AI lead. For each, a research agent identified the person or people whose primary job is AI, data science or data at the firm. It then built a profile from public professional sources: firm bios, press releases, conference and podcast pages, published articles and public LinkedIn pages. A second, independent agent reviewed every record against its sources and changed at least one field in 187 of the 190 profiles. It removed details it couldn’t tie to a source, downgraded overstated titles and rejected people who weren’t really AI leads. Our first post’s data listed 142 firms with a dedicated AI lead. The research found no current lead at four of them, and a re-check on October 6 downgraded seven more, leaving 131 firms with a verified AI lead.

We recorded only professional facts. We collected no ages, family details, locations beyond work city, gender or ethnicity.

Figures on seniority, background and remit count the most senior AI person at each firm. Several fields are unknown for a quarter or more of leads. Unless stated otherwise, percentages are of those with a known value.