Algorithmic Administrative Justice

Abstract

At the same time that artificial intelligence and machine learning systems are deployed with increasing frequency and success in the private sector, governments around the world are increasingly looking to harness the power of these digital tools to improve a variety of governmental functions, including sorting mail, identifying hazardous chemicals, uncovering securities and tax fraud, and improving traffic flow in congested cities. With time, algorithms will play a much larger role in assisting—or even replacing—humans involved in governmental tasks. This article assesses the range of legal, ethical, and policy concerns implicated by governmental use of algorithmic tools. Although machine-learning algorithms and other automated tools present important challenges for government related to accountability, procedural justice, transparency, privacy, and equality, the issues presented are not qualitatively distinct from the government’s use of other complex analytic tools. Ultimately, existing legal principles should prove to be no intrinsic or insurmountable obstacle to the responsible deployment of artificial intelligence. Yet to help ensure that artificial intelligence is used responsibly, public administrators, elected officials, and concerned citizens must remain vigilant in their use of such digital tools and see that machine-learning systems are ultimately deployed by governments in a manner consistent with both sound ethical judgment and sufficient empathy for those affected by these systems.