Legora is taking aim at one of the most difficult problems in legal artificial intelligence: determining not only what the law says, but whether it remains current, authoritative and applicable.
The legal AI company has unveiled a new research foundation that combines comprehensive legal data, a structured ontology of law and an AI-native citator within its platform.
The system is intended to help AI navigate the relationships among cases, statutes, regulations and other authorities. Legora says it will account for factors such as the hierarchy of courts, whether an authority is binding, how sources relate to individual legal propositions and whether subsequent developments have changed the law.
Those distinctions remain a major weakness for general-purpose AI systems. A model might cite an outdated rule, mistake a dissent for a holding or give the same weight to a regulatory decision and a Supreme Court judgment.
Legora says it has catalogued more than 50 ways AI can fail when conducting legal research.
Its response is to organize legal materials before an AI agent begins reasoning across them. The resulting ontology converts large collections of documents into structured data, while the citator helps establish how authorities have been interpreted, amended or displaced over time.
The technology draws on capabilities acquired through Legora’s purchases of legal research companies Qura and Wexler. The company has also assembled a team of lawyer-editors with experience developing traditional legal navigation tools to establish standards, address difficult cases and oversee quality control.
Building the underlying collection presents a separate challenge. Legora is working to bring together different types of legal data from more than 100 countries, partnering with publishers where possible and obtaining material directly from primary sources when commercial feeds are unavailable.
The company argues that combining the data and navigation layers within the same AI platform will allow lawyers to research, review and draft without moving between disconnected systems.
The new research infrastructure marks another expansion of Legora’s ambitions following a year of rapid growth and acquisitions.
As LegalTech.ca previously reported, Legora acquired Vancouver-based legal AI startup Walter in March. Founded by Ryan Wilson, Jon Conlin and Karl Campbell, Walter built agentic technology designed to work within lawyers’ existing workflows and developed its products alongside Canadian firms including Fasken and McCarthy Tétrault.
Walter’s team joined Legora to help pursue a broader vision of AI agents capable of handling legal work from beginning to end.
Legora subsequently opened its first Canadian office in Toronto, deepening its presence in a market where customers include McMillan LLP and Stewart McKelvey.
According to Legora, its platform is now used by more than 100,000 legal professionals across more than 1,800 law firms and in-house legal teams in over 50 markets.
The company’s new legal ontology is currently in limited beta and is expected to become generally available in the fourth quarter of 2026.