Failing at Face Value: The Effect of Biased Facial Recognition Technology on Racial Discrimination in Criminal Justice

Abstract

Recent years have seen a rise in the development of technological innovations and their implementation in various industries. Specifically, law enforcement agencies across the United States have partnered with technology companies to deploy facial recognition algorithms in the identification and prosecution of criminal suspects. Yet there is concern that law enforcement’s use of facial recognition algorithms based on biased mugshot data pools can lead to criminalizing innocent civilians. Prominent theories including intersection theory, instrumentalization theory, and Alvarado’s theory were analyzed to review arguments that justify concern. We find that intersection theory is supported by empirical evidence that women of color are put at the greatest disadvantage from technological bias; instrumentalization theory is supported by examples of both positive and negative implementations of facial recognition technology, and Alvarado’s theory further suggests the possible reinforcement of existing biases by these poor applications of technology.