Verification in Legal AI Is a Design Problem
Lawyers remain responsible. Legal AI should make its work easy to check before they sign.
Blog
Insights, updates, and stories from our team
Mary Technology hires litigator Catherine Thompson to lead Mary Legal in the U.S
Thompson joins from Blevans, Itzkowitz & Cantrell to own the standard Mary’s work must meet before an American lawyer relies on it.
The factual record is becoming infrastructure
As legal AI gets better, the record underneath the work matters more.
When legal AI changes information you have already verified
A new study of frontier models shows how legal AI can alter correct source information after retrieval, and why the system around the model can materially change the result.
Legal AI vendors should share the verification burden
Lawyers remain accountable for the work they use. But vendors should be accountable for making that work practical to verify.
The Two Clocks, in Litigation
The capability clock is moving faster than the institutional one. In litigation, the gap is adversarial.
What Is the Factual Record of a Litigation Matter?
A practical definition of the factual record in litigation, how it differs from documents and chronologies, and what it should preserve as a matter changes.
How to Test Legal AI on a Matter Your Firm Already Knows
A controlled way to test legal AI for source accuracy, material omissions, correction handling and the time required to reach usable work.
eDiscovery, Document Management and the Factual Record: Which System Does What?
A practical map of the roles played by document management, eDiscovery, factual-record systems and drafting tools across a litigation matter.
What MCP Means for Legal AI: Connection, Context and the Factual Record
A practical explanation of MCP clients and servers, what the protocol connects, and why legal context, permissions and reviewed factual state still need to be designed.
The Perfect AI Is Actually a Combination of Tools: A Litigation Guide
A practical guide to matching legal research, eDiscovery, factual review and drafting tools to the task each performs best.
Webinar: AI in Litigation: The Promise, the Pitfalls, and the Sanctions
A recap of our July 2026 panel with litigators Seth Goldstein and Catherine Thompson on the practical value, verification risks and professional obligations surrounding AI in litigation.
When a Lawyer Corrects the AI, Where Does the Correction Go?
A practical model for storing lawyer corrections as governed matter state rather than letting them disappear inside one AI conversation.
When the Record Is Incomplete, Legal AI Should Say So
How legal AI should handle insufficient evidence, conflicting accounts, missing documents and unread material without producing a complete-looking answer.
When a Real Source Still Produces a Wrong Legal AI Answer
A practical taxonomy for legal AI errors that survive a basic source check, including misquotation, mischaracterisation, transformation and incomplete coverage.
The Real ROI of Legal AI: Measure Time to Usable Work
A practical method for measuring legal AI ROI across generation, review, correction, omissions, implementation and reuse.
A Permission Model for Legal AI Agents: Read, Propose and Act
A practical permission model for legal AI agents, separating access to information from the authority to change a matter or act outside the firm.
AI for Commercial Litigation: Building the Factual Record Across a Large Matter
How a persistent, source-linked factual record can support commercial litigation from early case assessment through discovery, witness preparation and hearing.
Getting Started (Safely): Six Tips for Attorneys
A practical six-step framework for law firms introducing generative AI, covering approved tools, client data, closed-matter testing, source verification and human accountability.
The 7 Deadly Sins in Family Law
Emily Lonsdale and family lawyer Catherine A. Thompson examine how grievance, coercive control and fact chaos can intensify conflict in family-law proceedings.
Checking the AI is not enough
Why checking that a source exists can still miss factual omissions, misread dates and incomplete matter coverage, and what a stronger verification process should test.
Mary now integrates with Actionstep
If Actionstep is your Practice Management System, you can now upload documents straight into the Mary platform.
Cognitive Surrender: The Hidden Risk in Traditional AI-Assisted Litigation Workflows
Research on cognitive surrender shows how plausible AI answers can displace independent reasoning under time and complexity pressure. This article examines the risk in litigation and the controls that can reduce it.
Mary adds custom chronologies, bank statement analysis and template-based drafting
An overview of Mary's custom chronologies, bank statement analysis and template-based drafting, with results kept traceable to the underlying evidence.
Document Overload? How AI Fact Management Cuts Review Time
How legal teams can reduce repeat document review by structuring facts, sources and gaps across a litigation matter.
What Legal Professionals Need to Know About AI and How Not To Use It
A practical guide to task selection, client data, source verification, closed-matter testing and lawyer accountability.
Avoiding the Hype Trap in Legal Tech: What to Ask Your Vendor
A due-diligence checklist for law firms evaluating legal AI vendors, including testing, omissions, source fidelity, security and implementation.
How Legal Ops Teams Can Drive Tech Adoption
A practical framework for small legal-technology pilots with a defined task, known matter, committed users and measurable results.
The Complete Guide to Legal Fact Management
An introduction to legal fact management and how litigation teams turn scattered documents into a factual record that remains traceable to the evidence.
Manual vs AI Chronology Drafting: Process, Time and Verification
A practical comparison of manual and AI-assisted chronology drafting, including source review, omissions and the work lawyers still need to perform.