This field is required or invalid. Please provide a valid input.
This field is required or invalid. Please provide a valid input.
This field is required or invalid. Please provide a valid input.
This field is required or invalid. Please provide a valid input.
This field is required or invalid. Please provide a valid input.
Mode of Participation: *
This field is required or invalid. Please provide a valid input.
Mailing List Registration: *
This field is required or invalid. Please provide a valid input.
MPI Privacy Policy: *
This field is required or invalid. Please provide a valid input.
This field is required or invalid. Please provide a valid input.

Fields marked with * are required.

Artificial Justice Speaker Series - Virtual Lecture with Neel Guha

 

"An Institutional View of Legal AI Benchmarking"

 Wednesday, 13 May 2026, 4:00 p.m. - 5:00 p.m. (CEST)

 

About the Speaker:

Neel Guha is a PhD candidate in Computer Science at Stanford University and a graduate of Stanford Law School. His research sits at the intersection of artificial intelligence and law, focusing on how machine learning systems can improve legal practice and expand access to justice, as well as how legal institutions should adapt to the growing use of AI. His work on benchmarking and evaluating AI systems for legal applications has been widely adopted in the legal technology sector, and he has published extensively on topics including AI governance, liability, and regulatory design.

 

About the Topic:

Despite growing enthusiasm for the use of artificial intelligence in legal practice, there remains limited public information about the performance and risks of widely deployed systems. Recent findings have highlighted persistent issues such as “hallucinations,” where models generate inaccurate or fabricated legal information, raising concerns at the highest levels of the judiciary. This contribution argues for the importance of public benchmarking in legal AI. It examines the relative opacity of the legal AI ecosystem, the institutional challenges involved in designing meaningful benchmarks, and the risks of capture or misuse. Rather than proposing a single solution, it explores how different benchmarking approaches must be tailored to varying resource constraints, emphasizing the need for carefully designed institutions to ensure transparency, reliability, and accountability in legal AI.

 

About the Speaker Series:

The Artificial Justice Speaker Series features guests working at the intersection between law, computer science, and the humanities. Neither technical nor juristic knowledge is a prerequisite for participation—the Series is aimed at anyone with an interest in critical and interdisciplinary perspectives on “Law and AI”. The event takes place on Zoom and is scheduled to last one hour.

Max Planck Institute for Comparative and International Private Law