Algorithmic Bias and Racial Disparities in AI-Driven Criminal Justice Systems

Book Title: Computational Criminology: AI Applications in Forensic Science and Criminal Justice

Editors: Dr. Xavier Louis, Dr. Surbhi Girdhar, Ms. Aswathi Chandran Nair, Mr. Ravi Kumar, and Ms. Nandini Katare

Chapter: 26

DOI: https://doi.org/10.59646/704/26

Author: Shreya Singh Parihar

Abstract

The United States incarcerates its population at the highest rate in the world, and this mass incarceration is profoundly racially unequal: Black Americans are incarcerated at more than five times the rate of white Americans; Latino Americans at approximately twice the white rate.[1] These disparities are not adequately explained by differential crime commission rates and instead reflect the cumulative operation of racially differentiated points of contact with the criminal justice system. At the front end, over-policing in Black and Latino communities increases the frequency of surveillance, stops, searches, and arrests, thereby generating a larger volume of recorded criminal justice interactions independent of actual differences in offending behaviour. These initial contact disparities are then amplified through prosecutorial discretion in charging decisions and plea bargaining practices, where similarly situated defendants may receive different charges, plea offers, or bargaining leverage depending on jurisdictional context and case-specific interpretations that are themselves shaped by institutional patterns and resource asymmetries.


[1]The Sentencing Project, Report to the United Nations on Racial Disparities in the US Criminal Justice System (2018) 3.