Manipulating Risk: Immigration Detention Through Automation

The U.S. Department of Homeland Security arrests more than 500,000 migrants per year and detains more than 350,000 of them through Immigration and Customs Enforcement (ICE). Since 2012, ICE has relied on an automated Risk Classification Assessment (RCA) system to recommend who to detain and who to release. This article is the first to make that system’s methodology public. While purporting to base these recommendations on indicia of flight risk and risk to public safety, FOIA results unmask an algorithm driven by politics. By linking detention to enforcement priority, rather than risk, the RCA lost its underpinning in the constitution. In addition, compromises in its methodology thwarted the program’s ability to deliver the harm reduction, transparency, and uniformity it promised. Ultimately, our data and analysis reveal that manipulation of the RCA resulted in automated detention recommendations for hundreds of thousands of vulnerable people in violation of the constitution. The RCA’s fate was not inevitable, however. To limit the harms caused by immigration detention and restore policy and legal norms, DHS’s risk tool must rely on established indicators of flight and dangerousness, its logic should be reevaluated based on outcomes, and its assessments must be subject to review and challenge by immigration judges, advocates, and the people it detains.