Governing Under Moral Uncertainty: Institutional Design in the Age of Digital Minds

Abstract

As artificial intelligence systems grow increasingly sophisticated, public debate has begun to shift from questions of impact to questions of status. Much of contemporary AI governance focuses on harms such as bias, misinformation, privacy breaches, and labour displacement. A more complex institutional challenge may now be emerging: the possibility that the moral status of advanced systems becomes socially contested. This paper argues that the primary governance risk in such a scenario is not machine consciousness itself, but institutional instability under persistent uncertainty. When classification becomes unclear, decision-making frameworks designed for stable ontologies may struggle to function coherently. Rather than waiting for definitive scientific resolution, institutions must develop governance models capable of operating responsibly amid sustained ambiguity.