The Hidden Truth About Algorithms: Injustice in the Criminal Justice System

JurisdictionUnited States,Federal
AuthorBy Tangela Terry, Esq., and Ashley A. Fan
Publication year2022
CitationVol. 22 No. 1
THE HIDDEN TRUTH ABOUT ALGORITHMS: INJUSTICE IN THE CRIMINAL JUSTICE SYSTEM

By Tangela Terry, Esq.,1 and Ashley A. Fan2

INTRODUCTION

Algorithms have become a part of our everyday lives. From matching us with potential partners on dating apps, to their use of in the employment process, algorithms are all around us. While algorithms may seem beneficial to some, algorithms in the criminal justice system tend to disproportionately impact people of color. The truth about algorithms is hidden and more individuals ought to know about the truth in order to bring awareness to this important, modern-day issue.

RISK ASSESSMENT TOOLS AND COMPAS

One common use of algorithms in the criminal justice system is with risk assessment tools. Oftentimes, such algorithms are racially discriminative. Risk assessment tools are used in almost every state and many courts utilize them pretrial.3 The purpose of risk assessment tools is to predict a defendant's future behavior and quantify their overall risk by assigning them a risk score number.4 For instance, a black defendant who had four juvenile misdemeanors and no subsequent offenses was assigned a score of an eight and labeled as "high risk." Whereas a white defendant who was convicted of two armed robberies, one attempted armed robbery, and a subsequent offense of grand theft was assigned a score of a three and labeled as "low risk."5These risk assessment tools will consider factors such as a defendant's socioeconomic status, family background, and neighborhood crime given their zip code.6

ProPublica, a non-profit organization based out of New York City, reported that the Correctional Offender Management Profiling for Alternative Sanctions (COMPAS) is racially discriminative yet is still popular in courtrooms across the United States.7 The parent company of COMPAS, Northpointe (currently rebranded as Equivant), argued that overall, black defendants have a higher risk of reoffending after they have been arrested.8 Therefore, Northpointe argues that because of the higher baseline risk of black defendants, black defendants simply have higher risk scores.9 Northpointe's argument lacks evidence and is far from the truth because it artificially negatively categorizes these individuals for no real reason. Despite the racially discriminative issues associated with COMPAS, COMPAS is one of the most widely used systems throughout the United States.10 Even more alarming, most jurisdictions have adopted risk assessment algorithms, such as COMPAS, without first testing their validity and recognizing the hidden truth about algorithms.11

Of what does COMPAS consist? COMPAS consists of 137 questions directly answered by the defendant or pulled from their criminal record.12 Race is not a question asked by COMPAS.13 Rather, COMPAS consists of questions such as "was one of your parents sent to jail?" and agree or disagree

[Page 11]

to statements such as "does a hungry person have the right to steal?"14

Through the use of COM PAS, black defendants are incorrectly labeled as high risk to commit a future crime two times as often as a white defendant.15 Furthermore, COMPAS is oftentimes incorrect. The recidivism rate is only correct 61% of the time and only 20% of the defendants predicted to commit a violent crime committed one.16 It is scary to think that an algorithm, such as COMPAS, may be the sole basis that a for making a decision.

VERNON PRATER BRISHA BORDEN
Prior Offenses 2 armed robberies, 1 attempted armed robbery Prior Offenses 4 juvenile misdemeanors
Subsequent Offenses 1 grand theft Subsequent Offenses None
LOW RISK 3 HIGH RISK 8

THE IMPACT OF ALGORITHMS ON INDIVIDUALS' LIVES

There are numerous stories of individuals whose lives have been...

To continue reading

Request your trial

VLEX uses login cookies to provide you with a better browsing experience. If you click on 'Accept' or continue browsing this site we consider that you accept our cookie policy. ACCEPT