Kirkland & Ellis, the highest-grossing law firm in the world, has committed roughly $500 million over the next three to four years to build its own artificial intelligence systems in partnership with Palantir Technologies, rather than relying on the off-the-shelf legal AI tools most of its competitors are buying. About $100 million of that is going out the door this year alone.
The firm’s first product out of the partnership automates private equity fund formation, the paperwork-heavy process of drafting fund documents, side letters, investor terms and compliance filings that traditionally eats up thousands of billable associate hours. Kirkland says more than 1,000 of its lawyers are expected to use the resulting platform, which is built around the firm’s own institutional playbooks rather than generic legal templates.
Why Kirkland is going its own way
Most large firms have leaned on commercial legal AI products built by outside vendors and trained on general legal text. Kirkland’s pitch is that a tool trained on its own decades of deal documents, negotiating positions and client relationships is worth more to a private equity client than a generic AI wrapper, and that the $500 million buys the firm a proprietary edge its competitors can’t simply subscribe to.
Palantir’s role is less about the underlying language model and more about what the company calls the connective tissue: linking documents, client data, deal terms and compliance rules into a single system that can be queried and acted on. That’s the same ontology-based approach Palantir sells to defense and intelligence clients, now being pointed at fund paperwork instead of battlefield logistics.
What it means for the legal industry
A $500 million bet from a single law firm is a signal to the rest of Big Law that AI spending is no longer a side project. Firms that don’t have Kirkland’s revenue, and few do, are going to have a harder time matching that kind of proprietary investment, which could accelerate a split between firms that can afford to build custom AI infrastructure and firms stuck buying whatever’s on the shelf.
It also raises the obvious question hovering over every AI deployment in a white-collar industry: what happens to the army of junior associates whose job has historically been assembling exactly this kind of paperwork. Kirkland has framed the platform as freeing lawyers for higher-value work, but a tool built specifically to automate document preparation at scale is, by design, built to need fewer people doing that preparation by hand.
Sources: Kirkland & Ellis and Bloomberg Law.
