Artificial intelligence as a public problem: a multidimensional agenda for public administration

Abstract

The recent adoption of artificial intelligence (AI) systems in the public sector has often been framed as a matter of administrative efficiency and technological innovation. Such a perspective, however, tends to overlook broader institutional, distributive, and organizational implications associated with the digital transformation of the State. This article conceptualizes AI as a multidimensional public problem whose effects cut across state capacities, regulatory arrangements, and rights protection. Drawing on an integrative literature review, we develop an analytical framework structured around five cross-cutting dimensions – governance and regulation, inequalities and biases, state capacity, digital sovereignty, and work – articulated with key public policy sectors. Based on this framework, we propose a multidimensional matrix and a research agenda that systematize recurring dilemmas and guide future empirical investigations. The study contributes to Public Administration scholarship by integrating fragmented debates on AI and repositioning technology as an issue of governance, accountability, and social justice, offering conceptual guidance for more responsible and democratic digital policies.