Inteligencia Artificial en la administración pública: ambigüedad y elección de sistemas de IA y desafíos de gobernanza digital

Abstract

As public administrations embrace artificial intelligence (AI) it is noted that this transition has the potential to transform public service and public policies by offering rapid changes in decision making and service delivery. However, a recent series of critiques have pointed out the problematic aspects of integrating AI systems into public administration, noting some problematic outcomes in terms of justice and values. The argument provided here is that any public administration adopting AI systems must consider and address the ambiguities and uncertainties surrounding two key dimensions: the algorithms’ outcomes and how public managers make decisions for and about the design of AI systems. This article points out the need to design institutions that immerse themselves in understanding the nuances, details, and potential outcomes of AI governance for public administration. Such institutions would reconcile consequentialist logic with a logic of appropriateness to help navigate and mediate ambiguities and uncertainties.