Legora, the agentic operating system for legal professionals, is putting a new foundation under legal research in its platform: comprehensive data built on a full ontology of law and an AI-native citator. 

A large part of legal work begins with research: establishing what the law is and whether it still applies. It is also the step where AI has proved hardest to trust. Without a mapped structure of how legal authorities rank and relate, AI can’t fully understand every legal proposition and how it fits into the law as a whole.

“We are building the operating system that agents run on inside legal organizations, and legal research is a critical part of that,” said Max Junestrand, CEO and co-founder, Legora. “It is also the hardest thing in legal AI.”

Solving legal research requires two things: access to the law, and the ability to navigate it. Legora's new legal research will combine access to comprehensive data with AI-native navigation inside the platform Legora customers already use. And all of this comes with the confidence that the authority behind the data is current and binding.

“We don’t believe our clients should have to choose between the best data and the best technology,” said Melanie Brown, Head of Legal Data, Legora. “The frustration I hear most from clients is that they cannot access the data they already pay publishers for in the AI platforms they want to work in.”

Where Legora can partner with publishers, it does. Where a partnership is not on offer, it goes to the source directly, through manual requests, physical scanning, or whatever else the jurisdiction requires.

“It’s a different problem in every country,” said Brown. “In the US, access to citable case law is constrained by the companies that own the official reporters and won't sell a digital feed. In Germany, the challenges are digitization and anonymization. But there is always a way, and we’re doing whatever it takes to bring together every data type in over 100 countries around the world. We are bringing all the world’s legal data into Legora.” 

Data access alone is not enough, however. No agent can reason across hundreds of millions of legal documents. In a single matter, it cannot reliably reason across even a few thousand. The selection of sources to be used needs to be made before the reasoning starts, which means extracting the facts and the necessary relationships between documents, up front.

Legora is therefore building a full ontology of the law and an AI-native citator. Legora does this with proprietary technology from its acquisitions of Qura and Wexler, compressing corpora of thousands to hundreds of millions of documents into structured data that an agent actually can use to provide reliable output. 

“We have cataloged more than 50 distinct ways AI fails at legal research,” said Arvid Winterfeldt, who leads Legal Research at Legora and was co-founder and CEO of Qura, the legal research company Legora acquired earlier this year. “AI cites a rule and misses the amendment that changed it. It quotes a dissent as the holding. It puts a non-binding agency decision on a par with a Supreme Court case. A passage can look perfectly relevant in isolation and have been made obsolete by a later case.”

Building that navigation the traditional way took publishers 150 years and thousands of attorney editors, reading every opinion by hand and deciding what each paragraph meant.

“That's what AI changes,” said Brown. “For the first time, having someone, or something, read 60 million pages of case law is no longer prohibitive.”

To build its AI-native citator, Legora has assembled a team of attorney-editors who spent decades building publishers' navigation tools by hand. They define the standard, adjudicate the hard cases, and run ongoing quality control. 

The citator is enabled by building a full AI-native ontology of law that captures the relationships between sources, what each legal proposition means, the hierarchy of authority, temporal validity, and more.

Legora is used by more than 100,000 legal professionals at more than 1,800 law firms and in-house legal teams across over 50 markets, and this research foundation will sit under everything they do in the platform.

The ontology is currently in limited beta and will be generally available in Q4.