Affiliations: [a] Public Service and Administration Department, The Bush School of Government and Public Service, Texas A&M University, Texas, TX, USA | [b] Department of Public Administration and International Affairs, Maxwell School of Citizenship and Public Affairs, Syracuse University, Syracuse, NY, USA | [c] School of Public Administration, University of Nebraska-Omaha, Omaha, NE, USA
Correspondence:
[*]
Correspoding author: Justin Bullock, Public Service and Administration Department, The Bush School of Government and Public Service, Texas A&M University, Texas, TX, USA. E-mail: [email protected].
Abstract: This article examines the relationship between Artificial Intelligence (AI), discretion, and bureaucratic form in public organizations. We ask: How is the use of AI both changing and changed by the bureaucratic form of public organizations, and what effect does this have on the use of discretion? The diffusion of information and communication technologies (ICTs) has changed administrative behavior in public organizations. Recent advances in AI have led to its increasing use, but too little is known about the relationship between this distinct form of ICT and to both the exercise of discretion and bureaucratic form along the continuum from street- to system-levels. We articulate a theoretical framework that integrates work on the unique effects of AI on discretion and its relationship to task and organizational context with the theory of system-level bureaucracy. We use this framework to examine two strongly differing cases of public sector AI use: health insurance auditing, and policing. We find AI’s effect on discretion is nonlinear and nonmonotonic as a function of bureaucratic form. At the same time, the use of AI may act as an accelerant in transitioning organizations from street- and screen-level to system-level bureaucracies, even if these organizations previously resisted such changes.
