Large language models like ChatGPT present numerous opportunities in the realm of Law & Economics as a tool to assist in research. Specifically, these AI models have particular utility in their potential to automate hand-coding tasks for large datasets. In this project, we explore the feasibility of utilizing generalized large language models for the automation of classification, named entity recognition, and advanced language processing tasks in law-specific research. We do so by comparing the output of OpenAI’s GPT 3.5 and GPT 4 models in comparison to that of legal experts in analyzing aspects of Supreme Court cases.