AI for legal epidemiology

projects
epidemiology
law
Legal epidemiology requires a way to systematically understand what the law is, over time and across jurisdictions.
Published

November 12, 2025

Modified

December 26, 2025

Epidemiology starts with data

Overview

Legal epidemiology – understanding the effects of laws on health – requires an understanding of what the law is. But laws are spread across jurisdictions and different levels of government, and are framed in technical language that can be difficult to encode for analysis. Since 2024, I have been working with collaborators on a project to apply large language models and modern retrieval methods to make encoding of laws easier, focusing on the municipal codes and ordinances of hundreds of citites in the United States. This is part of a broader interest in making law more accessible.

Timeline

  1. Initial presentation to the COEP center at NYU Langone (March 28, 2024)

  2. Internal presentation with updated slides describing prototypes (Oct. 7, 2024).

  3. Magdalena Cerda et al., Stemming the Tide of the US Overdose Crisis: How Can We Leverage the Power of Data Science and Artificial Intelligence?, Milbank Q. 2025; 103(SI):0610 (June 10, 2025).

  4. Presentation at Society for Epidemiologic Research (SER) Annual Meeting (July 13, 2025).

Current version of code

Legiscope.

People

  • Samrachana Adhikari (NYU Langone)
  • Kimberly Villalobos Carballo (NYU)
  • Magdalena Cerda (NYU Langone)
  • Corey Davis (Network for Public Health Law)
  • Jaskiran Dhinsa (NYU Langone)
  • Charles DiMaggio (NYU Langone)
  • Theodore Hill (NYU Langone)
  • Spruha Joshi (University of Michigan)
  • Daniel Neill (NYU)
  • Pooja Shah (NYU Langone)
  • Utkarsh Srivastava