Policing by Proxy: Interrogating Big Tech\u27s Role in Law Enforcement

Predictive policing, sometimes referred to as data-driven or actuarial policing, is a method of policing that uses a risk-based approach to law enforcement. For-profit technology companies market proprietary risk assessment algorithms to law enforcement organizations as tools meant to proactively mitigate crime. Using data collected from a vast array of sources, both personal and public, police are able to predict the likelihood of criminal activity in a given area using these algorithms. Proponents claim that risk assessment tools have the potential to fight crime with unbiased accuracy and speed by predicting when, where, and whom to police by relying on mathematics and data analytics. Critics argue however, that historically marginalized communities have been, and continue to be marginalized due to the use of data that act as proxies for race and ethnicity in risk assessment algorithms. For this reason, members of these communities are disproportionately affected by predictive policing practices, which has led to their overrepresentation in our modern carceral system. This paper provides a framework for understanding the link between America’s historic racial hierarchy and the disproportionate policing and/or incarceration of communities of color, and calls into question the supposed unimpeachable nature of data analytics in policing