Every lawsuit is, at bottom, an argument about facts: who sent which email; what a witness said in March and whether it matches what they say in court; when a payment left an account and who approved it. In a large commercial dispute, those facts are buried in hundreds of thousands of emails, contracts, spreadsheets and transcripts, and finding them has traditionally meant teams of junior lawyers reading for weeks.
Wexler is a London software company that hands much of that reading to artificial intelligence (AI). Its platform, which it calls a “fact intelligence” system, pulls the facts out of a case file, links each one to the document it came from, arranges them into a timeline and points out where the evidence contradicts itself. Some of the world’s largest law firms use it, including Clifford Chance, Goodwin and Herbert Smith Freehills Kramer.
In July 2026 the Swedish legal AI company Legora bought Wexler, its fifth acquisition of the year, and the product is now part of Legora’s business. This article explains what the platform does, how its main features work and where it fits in the fast-growing market for legal AI.
From Founders In Residence To Legora
Wexler was founded in 2023 by Gregory Mostyn, its chief executive, and Kush Madlani, after the pair spent almost a year as founders in residence at Entrepreneurs First, the London programme that helps people find co-founders and start companies. The product launched in April 2024.
The company raised a $1.4M pre-seed round in December 2024, led by Myriad Venture Partners, and a $5.3M seed round in September 2025, led by Pear VC with Seedcamp and The LegalTech Fund. By the time Legora bought it, Wexler had offices in London and New York and had reported eightfold revenue growth over the previous 12 months.
Legora, which has raised $600M and was valued at $5.6B after a funding extension in May 2026, did not disclose the price.
“Every dispute and every investigation comes down to the facts,” said Max Junestrand, Legora’s chief executive and co-founder. Referring to case files that can run to more than a million documents, he called Wexler “the infrastructure built to handle that scale today”.
The Problem It Tackles
Litigation is the formal process of resolving a dispute through the courts, and arbitration is a private alternative to it. In both, each side has to build an accurate picture of what happened from the evidence. Much of that evidence, sometimes hundreds of thousands of documents, arrives through disclosure, or discovery in the United States (US), the process in which each party hands over relevant documents to the other.
Lawyers then have to turn that pile into a chronology, a dated list of events that shows how the dispute unfolded, and spot where a witness’s account, a contract and an email chain do not agree.
“Just getting to the facts is really complex and very time consuming,” said Charlie Morgan, a partner at Herbert Smith Freehills Kramer.
Mostyn has described Wexler as “a system that thinks like a litigator but operates at machine scale and speed.”
The company pitches it not as a replacement for the lawyers who argue a case but as a way to take over the slow first pass through the evidence so that they start from an organised factual record.
Inside The Platform
A legal team uploads a case file of up to 500,000 documents at a time, according to the company, in languages including English, French, German, Dutch and Italian. From there, the platform works through a set of connected features.
1. Fact Extraction
Wexler starts by turning documents into facts. It reads emails, witness transcripts, spreadsheets and reports and pulls out events, dates and assertions, such as a meeting taking place, a price being agreed or a person denying knowledge of something. Each fact is tagged, made searchable and linked back to the exact source document, so a lawyer can click through to check it.
2. Automatic Chronologies
The extracted facts are then arranged into a timeline of the case, with a note of why each event matters to the dispute. Building a chronology by hand is one of the most time-consuming jobs in litigation, and Wexler says work that once took days now takes about an hour.
3. Inconsistency Detection
The platform compares facts against each other and flags where they clash: two witnesses giving different dates for the same meeting, a claim that no document supports or a contract term that does not match an email. It also marks gaps and overlaps in the timeline. In one international arbitration described on Wexler’s website, a global firm went into the hearing relying on inconsistencies the platform had found.
4. Kim, The Case Assistant
Kim is Wexler’s AI assistant. Lawyers can ask it questions about the case in plain English, such as when a director first learned of a defect. It answers using only the documents uploaded, with each answer linked to its sources and given a confidence score. Kim can also carry out tasks in several steps and draft correspondence, memos, reports and court applications based on the case material.
5. Wexler Real-Time
Wexler Real-Time works while evidence is being given. During a court hearing or a deposition (the US practice of questioning a witness under oath before trial), it compares what the witness says against the uploaded documents and flags statements that conflict with them as they happen. Wexler launched it with its seed round in September 2025.
Accuracy And Trust
One of the biggest worries about AI in law is hallucination, when a model produces something that sounds right but is false, such as a quote nobody said or a case that does not exist. Courts in several countries have criticised lawyers for filing AI-generated material that turned out to be wrong, and the risk is similar to the one that has led Google to tell publishers to check AI drafts by hand.
Wexler’s answer is grounding: every fact and every answer from Kim is tied to a specific document, and Kim draws only on the user’s own files. A legal director at a national law firm said in Wexler’s published testimonials that the hallucination rate in the firm’s testing was below 0.1%, and that the tool performed at least on a par with human review. That figure comes from one firm’s tests rather than an independent study.
At one firm in the AmLaw 100, the list of the largest US law firms by revenue compiled by The American Lawyer magazine, Wexler says lawyers worked through more than 250,000 pages with an average time saving of 90%. In its seed announcement, it said customers had cut manual review by at least 75%.
Security For Sensitive Files
Case files are among the most confidential documents a company holds, so security sits at the centre of how legal software is sold. Wexler says it holds ISO 27001 certification, the international standard for managing information security, and a System and Organization Controls (SOC) 2 Type II report, an independent audit showing its security controls worked over a period of time. It also says it complies with the General Data Protection Regulation (GDPR), the European Union’s data protection law.
Each user’s data is protected with its own encryption keys, personal information can be masked, and firms can sign in through single sign-on (SSO), using the same login they use for other work systems. Data can be stored in the United Kingdom (UK) or the US, or elsewhere on request, and private cloud set-ups are available for firms that want their data kept apart from other customers’ data.
Who Uses It
Wexler is built for disputes teams at law firms, in-house legal departments and investigators and is aimed at firms with 100 or more lawyers. More than 70% of its users were in the US at the time of its seed round, with growth in the UK, Europe, Canada and Australia.
Clifford Chance has rolled it out across its dispute resolution practice, and Burges Salmon and Addleshaw Goddard are among its other customers. “It’s a product for litigators, made by litigators,” said Sarah McAtominey, a partner at Goodwin. Junior lawyers use it to organise documents and data, while senior lawyers use it to get quick updates on a case.
Wexler works in a crowded field, where legal AI companies across Europe are raising money to take on long-established legal information providers. Morning Tick reported when the legal AI startup Pandektes raised €13.5M to challenge the legal research provider LexisNexis. Wexler’s focus has been narrower: the facts of a dispute rather than legal research or contract drafting.
Where Wexler Goes Next
Wexler’s 18-strong team is joining Legora’s London engineering hub, and Mostyn has said the product will keep serving expert litigators, investigators and in-house teams. “Building the case brain has never been harder,” he said when the deal was announced.
For readers watching how AI is changing professional work, Wexler offers a clear example of a pattern in which the software does the reading, sorting and cross-checking at a scale no team could match, while the judgement about what the facts mean and how to argue them stays with the lawyers who carry the case.
Frequently Asked Questions
What Does Wexler Do?
Wexler is an AI platform for disputes and investigations. It reads a case file, pulls out the facts, links each one to its source document, flags where the evidence contradicts itself and builds a timeline of events.
Who Owns Wexler?
The Swedish legal AI company Legora bought Wexler in July 2026. The price was not disclosed, and Wexler’s team is joining Legora’s London engineering hub.
Who Founded Wexler And When?
Gregory Mostyn and Kush Madlani founded Wexler in London in 2023, after almost a year as founders in residence at Entrepreneur First. The product launched in April 2024.
What Is Kim?
Kim is Wexler’s AI assistant. It answers lawyers’ questions using only the documents in a case, links each answer to its sources with a confidence score and can draft memos, letters, reports and court applications.
What Is Wexler Real-Time?
Wexler Real-Time compares what a witness says during a court hearing or deposition with the documents in the case file, and flags statements that conflict with them as they are made.
How Many Documents Can Wexler Handle?
Wexler says it can take up to 500,000 documents in a single upload, in languages including English, French, German, Dutch and Italian.
Which Law Firms Use Wexler?
Named customers include Clifford Chance, Goodwin, Herbert Smith Freehills Kramer, Burges Salmon and Addleshaw Goddard. The company says more than 70% of its users were in the US at the time of its 2025 seed round.
How Does Wexler Protect Client Data?
Wexler says it holds ISO 27001 certification and a SOC 2 Type II report, gives each user separate encryption keys, can mask personal information and stores data in the UK, the US or elsewhere on request.
Does Wexler Replace Lawyers?
The company does not pitch it that way. Wexler is built to handle the first pass through the evidence, sorting and cross-checking documents at scale, while lawyers decide what the facts mean and how to argue the case.
