Research Project

Predictive policing technologies are algorithmic agents that can forecast when crime will occur in a particular area, the type of crime that will occur, and who is more likely to commit a crime. It was initially implemented by police departments to thwart acts of conscious or unconscious bias in human-based policing. However, there are concerns surrounding the effectiveness of these technologies, due to allegations of racial bias and a lack of neutral police contact embedded within the policing data. I conducted a case study on two cities that have utilized predictive policing within their policing methods (Los Angeles’s PredPol software and Chicago’s ShotSpotter program) to examine how these technologies have influenced the crime rate and the broader community attitudes of neighborhoods that have engaged with algorithmic policing resources. I used the National Institute of Standards and Technology’s (NIST) Artificial Intelligence Risk Management Framework (AI RMF) as a lens for which community-oriented perspectives to data-driven policing can be applied. The NIST AI RMF is a voluntarily produced set of guidelines meant to help manage risks associated with AI use, and introduce authenticity and transparency within the development and deployment of AI-based products. My research considers how the NIST AI RMF can incorporate a community-based approach by administering social support services to uplift and alleviate communities mistreated by algorithmic wrongs.