Institutionalized respect in the algorithmic state: On the consequences of artificial intelligence for citizen-state relations

Making Artificial Intelligence (AI) systems in government work for the benefit of citizens hinges on functional properties such as their performance and transparency. However, the proliferation of these systems also amounts to a transformative change through which citizens face an increasingly algorithmic state. This makes it important to shift the view from AI systems themselves toward the changing quality of the citizen-state relationship. This article highlights the importance of a social-relational dimension of government AI uses for understanding how they can be problematic even when producing material benefits for citizens. It combines social recognition theory with institutionalist arguments to develop an integrative account of how AI use can reduce the institutionalized expression of respect toward citizens – and how this differs by setting of AI deployment. The framework provides a coherent approach for conceptualizing both moral implications of government AI use and non-instrumental evaluative criteria regarding citizens’ perceptions of these uses.