This post was written with the support of the LegalQuants community, a group of lawyers who are building some pretty great things with AI. Curious? Check it out at https://www.legalquants.com/.
Over the past several days I’ve been digging into the Claude for Word add-in, and the headline finding surprised me. On document-intensive legal work
Somewhere around tool number five, I realized I had a problem. The kind where you keep saying “almost done” and then it’s 2 AM and you’re finalizing a React app that lets your students act as bots and deploy hallucinations against one another.
I teach 21st Century Lawyering at The Ohio State University Moritz College
This is Part 1 of a three-part series on AI hallucinations in legal research. Part 2 will examine hallucination detection tools, and Part 3 will provide a practical verification framework for lawyers.
You’ve heard about the lawyers who cited fake cases generated by ChatGPT. These stories have made headlines repeatedly, and we are now
Way back in 2023, I thought it was amazing how I could use generative AI to streamline my Thanksgiving prep: I gave it my recipes, and it gave me a schedule. It was a static list—a text document that told me when to put the turkey in, when to swap in the stuffing, and so
I’ve been watching my legal research students use AI and noticing a common pattern.
They typically go into an AI “Ask” feature in Lexis or Westlaw, get an answer, and then continue the conversation by asking more questions. This is exactly what the tools are designed to encourage.
The problem is that this process often
The impact of AI on varied aspects of our professional lives is covered regularly on this blog. It is reshaping legal research, education, and legal practice in ways that threaten to leave us behind if we fail to be proactive. It is why the Future of Law Libraries Initiative gathered professionals from academic, court, firm,
As anyone who has taught legal research knows, Boolean searching is a superpower. The ability to craft a precise query with terms and connectors is the difference between finding a needle in a haystack and finding nothing at all. But for newcomers, the syntax of ( ), !, /p, and /s can feel like learning
Debbie Ginsberg, Guest Blogger
Benchmarking should be simple, right? Come up with a set of criteria, run some tests, and compare the answers. But how do you benchmark a moving target like generative AI?
Over the past months, I’ve tested a sample legal question in various commercial LLMs (like ChatGPT and Google Gemini) and RAGs
If you follow me on LinkedIn or spoke with me at AALL, you’ve probably seen me teasing this project like it’s the season finale of a legal tech drama. Well, the wait is (almost) over — here’s your official sneak peek at our forthcoming interactive GenAI Legal Hallucination Tracker.
The People Behind the Tracker
First,
AI may have struck again with hallucinations. Yesterday evening, I was forwarded a quote from the case opinion of Shahid v. Esaam, 2025 Ga. App. LEXIS 299, at *3 [Ct App June 30, 2025, No. A25A0196]) released on June 30, 2025 by the Georgia Court of Appeals. (HT Mary Matuszak!)(link to official opinion, not