BigLaw Build vs. License vs. Buy — The Kirkland-Palantir Model and Its Imitators

The Kirkland & Ellis–Palantir partnership — a multi-year deal to build a proprietary platform using Palantir AIP for private equity fund formation, documentation, side letters, investor tracking, and compliance, expected to reach more than 1,000 Kirkland lawyers — has continued to reframe the law firm AI strategy conversation in the six weeks since its announcement, with a July 22 Legaltech News analysis describing it as a bellwether for firms that have the balance sheet to build custom AI pulling further ahead of the field. Kirkland's AI team now numbers approximately 180 engineers and data scientists, backed by a $500 million infrastructure investment. The DraftWise briefing (August 3) frames the strategic choice for other firms as build, license, or buy a vertical platform — and notes that no firm has yet published data showing that a homebuilt system has actually reduced costs or improved outcomes for a client, making the multi-year development investment "a bet on a promise, not a demonstrated return." For law firm leadership teams making AI strategy decisions in 2026, the Kirkland model is instructive in two ways: it is capital-intensive and talent-intensive in ways that are not replicable for most firms, and it is focused on a specific practice area (private equity) with high-volume, structured documentation workflows that are suited to AI automation. Firms outside the elite BigLaw tier evaluating their AI strategy need to anchor their build-vs.-buy calculus to specific practice area economics rather than general AI ambition.