Two research papers co-authored by Michigan Law professors have won awards from the University for their effective use of artificial intelligence technology.
“Canons of Statutory Interpretation at the Supreme Court, 1791-2024,” a forthcoming paper co-authored by Nina Mendelson, the Joseph L. Sax Collegiate Professor of Law and co-director of the Environmental and Energy Law Program, won a first-place Empowering Research with AI award.
In addition, “AI-Powered Lawyering: AI Reasoning Models, Retrieval Augmented Generation, and the Future of Legal Practice” won a third-place award. The paper was authored by J.J. Prescott, Henry King Ransom Professor of Law, co-director of the Empirical Legal Studies Center and co-director of the Program in Law and Economics; Patrick Barry, clinical assistant professor of law and director of Digital Academic Initiatives; and several colleagues at other institutions.
Sponsored by AI Institutes at Michigan, the recognition “honors individuals and teams whose innovative and rigorous use and assessment of AI advances knowledge or practice in their field.” The winning papers were chosen for the awards from more than 180 submissions, representing dozens of schools and colleges across all three U-M campuses.
Mendelson’s paper, written with Jonathan Choi of Washington University in St. Louis, examines the “rules of thumb” that judges use to interpret statutes—such as the rule of lenity, which instructs courts to interpret ambiguous criminal laws in favor of a defendant.
“Despite decades of scholarly attention, empirical study of how canons actually operate has been limited by the difficulty of hand-coding cases at scale,” Mendelson said. “Prior studies have been confined to specific canons or narrow time periods.”
Mendelson and Choi used large language models (LLM) to analyze the use of canons across every Supreme Court opinion from 1791 to 2024. Their findings “complicate conventional wisdom about canons of statutory interpretation in several ways,” the researchers wrote.
“This project demonstrates how AI can transform empirical legal scholarship. Tasks that would have taken teams of research assistants years to complete—reading, coding, and classifying canons across tens of thousands of opinions—become feasible with LLM-powered pipelines validated against human benchmarks.”
The paper will be published in the Yale Law Journal in 2027.
—————
An interdisciplinary approach to AI
Barry and Prescott’s research represents the first randomized control trial to test how AI tools affect the way lawyers perform core legal tasks. They found that next-generation AI tools can help lawyers
produce higher-quality legal work—including stronger analysis, clearer writing, better organization, and greater professionalism—while also improving efficiency.
The paper has been published in the Journal of Law & Empirical Analysis.
“There’s a lot of speculation about how AI will change the legal profession, but relatively little hard evidence. What this recognition highlights, to me, is the importance of grounding that conversation in data,” Prescott said. “Our study tries to answer a simple but consequential question—whether these tools actually make lawyers better at what they do—and the answer, at least for some tasks, is yes.”
Barry added, “One of the cool things about this project is how interdisciplinary it was. That mix of expertise is a useful reminder that artificial intelligence is the kind of subject that definitely benefits from something Michigan is really good at: using multiple perspectives and methodological approaches to study a complex problem.”
AI and related issues have become a major focus at Michigan Law. The school is meeting this moment with tools, resources, and academic rigor designed to forge the next generation of legal advocates, scholars, and policymakers. Specific efforts include the AI Law and Policy Clinic, the Artificial Intelligence Law and Policy Society, online courses, and more.
The school also recently launched an AI Advisory Council to bring industry insight and strategic guidance to help ensure that Michigan Law’s approach remains rigorous and responsive to a rapidly changing profession.
“It’s important for Michigan to recognize work like this because the legal profession is going to be shaped by AI whether we like it or not,” Prescott said. “Universities should be leading in figuring out what these tools actually do in practice—and how to use them responsibly. I’m thrilled Michigan is doing that.”
“Canons of Statutory Interpretation at the Supreme Court, 1791-2024,” a forthcoming paper co-authored by Nina Mendelson, the Joseph L. Sax Collegiate Professor of Law and co-director of the Environmental and Energy Law Program, won a first-place Empowering Research with AI award.
In addition, “AI-Powered Lawyering: AI Reasoning Models, Retrieval Augmented Generation, and the Future of Legal Practice” won a third-place award. The paper was authored by J.J. Prescott, Henry King Ransom Professor of Law, co-director of the Empirical Legal Studies Center and co-director of the Program in Law and Economics; Patrick Barry, clinical assistant professor of law and director of Digital Academic Initiatives; and several colleagues at other institutions.
Sponsored by AI Institutes at Michigan, the recognition “honors individuals and teams whose innovative and rigorous use and assessment of AI advances knowledge or practice in their field.” The winning papers were chosen for the awards from more than 180 submissions, representing dozens of schools and colleges across all three U-M campuses.
Mendelson’s paper, written with Jonathan Choi of Washington University in St. Louis, examines the “rules of thumb” that judges use to interpret statutes—such as the rule of lenity, which instructs courts to interpret ambiguous criminal laws in favor of a defendant.
“Despite decades of scholarly attention, empirical study of how canons actually operate has been limited by the difficulty of hand-coding cases at scale,” Mendelson said. “Prior studies have been confined to specific canons or narrow time periods.”
Mendelson and Choi used large language models (LLM) to analyze the use of canons across every Supreme Court opinion from 1791 to 2024. Their findings “complicate conventional wisdom about canons of statutory interpretation in several ways,” the researchers wrote.
“This project demonstrates how AI can transform empirical legal scholarship. Tasks that would have taken teams of research assistants years to complete—reading, coding, and classifying canons across tens of thousands of opinions—become feasible with LLM-powered pipelines validated against human benchmarks.”
The paper will be published in the Yale Law Journal in 2027.
—————
An interdisciplinary approach to AI
Barry and Prescott’s research represents the first randomized control trial to test how AI tools affect the way lawyers perform core legal tasks. They found that next-generation AI tools can help lawyers
produce higher-quality legal work—including stronger analysis, clearer writing, better organization, and greater professionalism—while also improving efficiency.
The paper has been published in the Journal of Law & Empirical Analysis.
“There’s a lot of speculation about how AI will change the legal profession, but relatively little hard evidence. What this recognition highlights, to me, is the importance of grounding that conversation in data,” Prescott said. “Our study tries to answer a simple but consequential question—whether these tools actually make lawyers better at what they do—and the answer, at least for some tasks, is yes.”
Barry added, “One of the cool things about this project is how interdisciplinary it was. That mix of expertise is a useful reminder that artificial intelligence is the kind of subject that definitely benefits from something Michigan is really good at: using multiple perspectives and methodological approaches to study a complex problem.”
AI and related issues have become a major focus at Michigan Law. The school is meeting this moment with tools, resources, and academic rigor designed to forge the next generation of legal advocates, scholars, and policymakers. Specific efforts include the AI Law and Policy Clinic, the Artificial Intelligence Law and Policy Society, online courses, and more.
The school also recently launched an AI Advisory Council to bring industry insight and strategic guidance to help ensure that Michigan Law’s approach remains rigorous and responsive to a rapidly changing profession.
“It’s important for Michigan to recognize work like this because the legal profession is going to be shaped by AI whether we like it or not,” Prescott said. “Universities should be leading in figuring out what these tools actually do in practice—and how to use them responsibly. I’m thrilled Michigan is doing that.”
––––––––––––––––––––
Subscribe to the Legal News!
https://legalnews.com/Home/Subscription
Full access to public notices, articles, columns, archives, statistics, calendar and more
Day Pass Only $4.95!
One-County $80/year
Three-County & Full Pass also available
