Q&A: Kirkland & Ellis on AI and its deal with Palantir

Q&A: Kirkland & Ellis on AI and its deal with Palantir

Madeline Shi, Esther Luz

Sat, September 12, 2026 at 12:04 AM GMT+3 6 min read

• PLTR +0.83%

Private equity advisers are racing to leverage AI and automate the services they provide to fund managers and their investors. Law firms, such as Kirkland & Ellis, have been among the most active adopters.

The dominant name in private fund formation and sponsor-backed M&A, the firm rolled out its own AI platform in June, built with Palantir and designed to reshape the repetitive, boilerplate elements of fund formation work, such as drafting limited partnership agreements, closing investor commitments and conducting compliance.

Earlier this year, Kirkland reportedly earmarked $500 million to develop a proprietary AI tool over three to four years, no small sum even for a law firm that generated $10.6 billion in annual revenue last year.

Erica Berthou

Erica Berthou, a partner at Kirkland and a member of the firm's executive committee, recently spoke with PitchBook about how AI could reshape the practice of law—including how it will change the economics of legal work—and how the adoption of AI may affect practitioners, especially the young talent.

The interview was edited for clarity and brevity.

PitchBook: Everyone in private equity talks about AI. What does adoption actually look like inside PE firms right now?

Erica Berthou: We spend a lot of time talking to in-house legal teams at our clients about how they can operate more effectively using AI. Private equity firms tend to have more off-the-shelf legal tools. They use them to summarize documents for quick review and to extract key terms — essentially, to ask a question and get an initial analysis. That does not replace inside or outside legal counsel or human judgment, but it certainly can speed up the initial analysis.

Kirkland launched its Fund Formation Engine with Palantir. What exactly did you build?

What we're building with Palantir is basically a bespoke fund formation engine that embeds all of our legal knowledge in the funds industry.

That includes our model document library, where we can prepare fund agreements using AI. We can respond to investor requests, drawing on potential accommodating solutions to address their concerns. And we can automatically track all the obligations and undertakings our clients make to their investors, ensuring they comply with everything they say they'll comply with.

The big differentiator versus off-the-shelf legal tools is that this is our legal knowledge. Off-the-shelf legal tools are trained on common knowledge and publicly available data.

Are LPs drawing any lines around how GPs use it?

I think limited partners, fund managers and law firms alike are actually very aligned in the use of AI for efficiency. Speed is an important factor in fundraising, for better analysis and more knowledge to inform human judgment. If you operate faster and have a better outcome, it's also more cost-efficient.

In reality, I think it's the speed of evolution that creates so many questions. Each individual has a steep learning curve when it comes to AI. Think about where we all individually were five years ago, where you were two years ago and where you are now.

The second aspect is organizational adoption. There are procedures to think through, including technological and legal safeguards around how you use it.

Is AI use showing up in fund documents, particularly regarding costs?

I haven't seen any real adoption of that in fund agreements as of yet. There is an appropriate cost that can be borne by a fund. You can almost think about [AI] like a document management system or using Microsoft Word, and how that is allocated today from a cost perspective. I don't think AI creates entirely novel issues from a cost perspective.

Sign up for The Daily Pitch newsletter Subscribe

GPs tell us they now run legal questions through AI before calling their lawyers, then come to counsel with sharper questions. How does that reshape the job, especially for junior associates?

It's hard to predict how this will look many years from now. But some of the simpler legal-administrative and data-extraction tasks — reviewing NDA upon NDA upon NDA — will be muted. Nobody became a better NDA expert doing 10,000, compared with 100.

I actually think younger lawyers will become more senior more quickly. There will be less repetitive work, and we will spend more time on delivering really thoughtful, well-counseled legal advice. We will spend more time on human interaction and less time just pinging emails back and forth.

I know a lot of people are nervous about this, but I think it is a really great development for our industry.

What do you expect from the associates you're hiring now?

They need to have open minds and open hearts. I don't expect a law student coming into our group today to be fully versed in using all kinds of AI or to have an engineering background.

But we do train all our lawyers when they join us, and we expect that, over time, they really get to AI adoption. The reality is that a human powered by AI, compared with a human who doesn't use any AI at all, will simply be more successful. They will progress faster.

Will AI end up writing your emails?

Let me answer that by asking you a cheeky question: Do you notice when somebody sends you an email drafted by AI or not?

I think AI will help prepare drafts of more complex emails. But at the end of the day, if you don't have the personal touch — a cute emoji or whatever your trademark is — how do you connect with people? Human connection is a really important differentiator.

You should be wary if you get an email from me with no Swedish English in it.

Will AI change how law firms charge for their work?

There is the possibility of a trend toward more project-based value billing, as opposed to a strict legal hour.

If you have a well-defined scope of work and experience working together, I think you can predict and agree upon an appropriate price for that work, as opposed to billing by the hour against a budget that may or may not be accurate.

For the right projects, I think both law firms and clients would welcome that alignment. I think fund investors in my area would like that too, because the predictability of legal spend is important for everyone.

Some clients will like it and some will not, so I don't want to overpredict how this is going to go. But we're keeping an open mind.

This article originally appeared on PitchBook News