Governing rapid technological change: Policy Delphi on the future of European AI governance

Abstract

Purpose Rapid advancements in artificial intelligence (AI) present unique challenges for policymakers seeking to govern the technology in an anticipatory way. This paper aims to explore the tensions between desirable and probable futures in European AI governance. Design/methodology/approach This research uses a two-round Policy Delphi study conducted in mid-2024 with a panel of 29 European AI experts from policymaking, academia and civil society. Findings This study identifies a desirability–probability gap in AI governance: desirable policy directions, such as robust citizen participation, are perceived as less probable and feasible. Furthermore, experts agreed that future-proofing AI regulation against technological change depends primarily on sustained institutional capacity to enforce it, rather than the legislation’s initial technical scope. Originality/value As one of the first Delphi studies focused specifically on AI governance, this research challenges the prevailing European Union assumptions on technology neutrality and regulatory sandboxes as future-proofing mechanisms. The desirability–probability gap demonstrates that the obstacle to preferred AI futures is not necessarily a lack of foresight, but of regulatory capacity.