Kirkland & Ellis, the world's highest-grossing law firm, committed $500mn in May to build its own AI platform, rejecting off-the-shelf tools available to competitors. The move signals a structural split in the legal AI market as elite firms seek proprietary technology for complex, high-stakes work while standard software handles routine tasks.
The global legal tech market is projected to reach about $70bn by 2033, up from roughly $31bn in 2026, according to Grand View Research. That growth is drawing both established suppliers like LexisNexis and Thomson Reuters, and newer entrants Harvey and Legora. But the largest firms are now charting a different course.
"If you are operating in a premium segment as legal adviser, you also need premium tools," said Sebastian Lach, partner at Hogan Lovells Cadwalader and co-chief executive of Eltemate, the firm's tech entity. Eltemate has built about 10 legal AI tools, including systems for litigation and regulatory tracking. "That translates into technology you basically can't buy off the shelf from a tech provider. The knowhow needs to be injected as part of the tool."
The buy-versus-build calculus
Jon Ballis, chair of Kirkland & Ellis, told the Financial Times that widely available AI tools were "raising the floor for everyone," but added that Kirkland had to do more because "we don't get hired for the floor." The firm's $500mn allocation marks the most aggressive bet yet on proprietary legal AI.
Freshfields, the UK-based magic circle firm, announced a deal with Anthropic in April to help build specialist legal tools that could eventually be sold to others. The firm pays Anthropic an undisclosed sum and gains early access to tech initiatives while deploying Claude AI across its business. Freshfields also maintains a 50-person tech lab in Berlin and continues using software from Harvey and Legora.
Alan Mason, global managing partner at Freshfields, said the firm is not tied to any single AI model. "We're not wedded to any particular model today, or technology, because we're fully expecting these things to evolve," he said. "If a client wants to use Harvey, we can deliver advice using Harvey. If a client wants to use Legora, we have the technology and the expertise to use Legora."
Goodwin Procter joined the trend in August, announcing it will spend about $25mn annually to buy and develop its own AI. The US-headquartered firm has already rolled out a venture financing tool.
A market in two tiers
Brian Tang, executive director of the University of Hong Kong's LITE Lab@HKU, estimates that about 20 percent of large corporate law firms worldwide are experimenting with building or personalizing their own legal AI. He sees a potential outsourcing market emerging for firms that want custom tools but lack the IT infrastructure to run them securely.
"If I'm a law firm and I don't want to be a software developer, I can now buy code or I can get access to some of these tools, maybe through open source tech," Tang said. "But if I don't know how to handle the cyber security stuff or the confidentiality stuff, maybe new service providers will supply this kind of outsourcing."
Standard AI software remains transformative for drafting contracts and conducting legal research. For many firms, the deeper question is not just whether to build or buy, but whether lawyers are fully using the tools already available. Max Junestrand, chief executive of Legora, said the gap between an average user and the most sophisticated one is substantial. "You can be extremely advanced and effective by really leveraging all of the different bells and whistles in the tool, versus being a novice."
Suppliers adapt to the shift
Harvey and Legora are not standing still as customers explore proprietary development. Some suppliers now help firms adapt their AI software to support new legal tasks or processes. Gabe Pereyra, president and co-founder of Harvey, said the Kirkland announcement has prompted more firms to ask: "How are we differentiating on AI?"
Still, Junestrand remains confident in Legora's position. "In the same way that I have no interest in developing my own law firm, I am not at all rattled by the idea that law firms would try to go and build the equivalent of Legora because, frankly, I don't think anyone ever could," he said.
For professionals navigating this shift, targeted training can bridge the gap between basic tool usage and the sophisticated application that firms now demand. The AI for Legal landscape is evolving quickly, and structured learning paths help paralegals and associates stay ahead of the curve. The AI Learning Path for Paralegals addresses exactly the document review and contract analysis skills that standard legal AI tools now handle, freeing time for the complex work that bespoke platforms target.
Why this matters for legal professionals
The split between standard and bespoke legal AI will reshape how lawyers work at different tiers of the market. For associates and paralegals at large firms, proficiency with off-the-shelf tools like Harvey and Legora is becoming a baseline expectation. For those at firms investing in proprietary platforms, understanding how to collaborate with tech teams and provide the legal expertise that custom tools require will become a distinct career advantage. The firms writing the biggest checks are betting that AI differentiation - not just adoption - will determine who wins premium mandates.
Your membership also unlocks:
