Paper title: GPT-4 Passes the Bar Exam and the Emerging Role of Natural Language Processing (NLP) within the Legal Domain

Abstract: This workshop will introduce the ideas in two papers—a survey paper on Natural Language Processing (NLP) & Law and an application paper using a Large Language Model (called GPT-4) on all elements of the Uniform Bar Exam—and relate those ideas, in conversation, to labor markets, legal innovation, and law-related behavior and practice. The presentation will begin with a general introduction to NLP and NLP & Law and a review of key findings from the NLP Survey Paper, where we offer a complete survey of the technical academic literature about applying NLP models within the legal domain. I will turn next to the GPT-4 Passes the Bar Exam Paper. I will discuss our analysis of GPT-4 on the Uniform Bar Exam (MBE, MEE, MPT) and discuss how to think about these findings. Time permitting, I will also briefly highlight some other application areas where NLP / Large Language Models (LLMs) are already (or could soon be) in use in law. With these preliminaries in place, I will identify potential labor-market, innovation-related, and other impacts that Large Language Models might have both within law and society.


Michigan's Law and Economics Workshop provides an opportunity for faculty and students from across the University to engage with cutting-edge law and economics research by leading scholars on a wide range of legal and policy topics.

The Winter 2023 workshop meets on Thursdays from 4:30-6:30 p.m. All workshops occur in person in Jeffries Hall 1025 and are open to the academic community from the University of Michigan and elsewhere.

Professor J.J. Prescott ( organizes the workshop. If you would like to receive workshop announcements, please contact Alex Wroble ( and ask to have your name added to the workshop’s email list.