BIMSA >
ICMRA Seminar Series
ICMRA Seminar Series
Judicial Systems as Objects of Scientific Inquiry: Learning, Uncertainty, and Institutional Dynamics
Judicial Systems as Objects of Scientific Inquiry: Learning, Uncertainty, and Institutional Dynamics
Organizer
Speaker
Nubia Regina Ventura
Time
Thursday, June 11, 2026 3:30 PM - 4:30 PM
Venue
A6-101
Online
Zoom 204 323 0165
(BIMSA)
Abstract
What can machine learning reveal about the hidden structure of law, and what does law reveal about the limits of machine learning? This talk presents a research agenda that approaches the Brazilian judiciary as a scientific object: a large-scale, evolving system whose behavior emerges from interactions among institutions, rules, people, and data. Drawing on an ongoing collaboration with the São Paulo Court of Justice (TJSP), the project combines empirical findings from classification, clustering, and segmentation models with a cross-disciplinary framework for understanding bias and risk in judicial AI, drawing from machine learning, legal theory, and the sociology of inequality. The long-term goal is the development of an empirical risk index for AI tools in judicial settings, and a book conceived as an interdisciplinary scientific contribution across mathematics, law, and social science. We believe judicial systems offer a rich and underexplored laboratory for formal modeling, and welcome collaborators who see in institutional and social complexity a productive frontier for their own research.