On LawNext Podcast: Can the AI of GPT-3.5 Pass the Bar Exam? We Find Out

Uncategorized
This post was originally published on this site

Since the December release of Open AI’s GPT-3.5 model, and the related ChatGPT, speculation has been rampant about how this next generation of artificial intelligence might upend the legal profession. But as others have been speculating, two legal scholars and scientists, Daniel Martin Katz and Michael Bommarito, put GPT 3.5 to the task, having it perform that most anxiety-inducing of tests along the path to becoming a lawyer – taking the bar exam. 

How did it do? Katz and Bommarito recently published the results in their article, GPT Takes the Bar Exam, and on this episode of LawNext, they join me to discuss why they did this experiment, how it turned out, and what it all means for the future of AI in law. 

Katz is professor of law at Chicago-Kent College of Law and academic director of both The Law Lab at Illinois Tech, Chicago-Kent College of Law and the Bucerius Center for Legal Technology & Data Science in Hamburg, Germany. He is cofounder and CSO of 273 Ventures, and formerly cofounded the legal AI company LexPredict, which was acquired by Elevate in 2018. 

Bommarito is cofounder and CEO of 273 Ventures and a serial entrepreneur and investor with over 20 years of experience in the financial, legal, and technology industries. A cofounder with Katz of LexPredict, he also co-founded Telly, an open source telemetry platform, and licens.io, an information security and compliance data company. He is an adjunct professor at Michigan State University College of Law.

Thank You To Our Sponsors

This episode of LawNext is generously made possible by our sponsors. We appreciate their support and hope you will check them out. 

If you enjoy listening to LawNext, please leave us a review wherever you listen to podcasts.