UK AI Governance at a Crossroads: Charting the Path Ahead

Abstract

Governance of Artificial Intelligence (AI) in the UK remains in its formative stages, guided by the policy objective of a “proportionate and pro-innovation regulatory framework” as set out in the White Paper (DSIT, 2023). Consistent with this framing, the UK has adopted a cross-sectoral, principles-based approach that seeks to balance regulatory flexibility with innovation while addressing AI-related risks. Rather than enacting comprehensive AI-specific legislation, the government relies on sectoral regulators to implement overarching principles - safety, transparency, fairness, accountability and contestability - through existing legal and institutional frameworks (DSIT, 2023).This trajectory is reflected in key initiatives such as the National AI Strategy (DSIT, 2022a), the establishment of the AI Safety Institute, and the AI Opportunities Action Plan (DSIT, 2025), collectively signalling a decentralised and innovation-friendly governance model. Nevertheless, whether this regulator-led approach should be maintained or replaced by a more uniform, horizontal legislative framework remains an increasingly contested question.To address this issue, this paper examines how AI is currently regulated in three sectors - healthcare, education and legal services - and evaluates the extent to which cross-sectoral principles are being operationalised consistently and effectively. It contrasts these practices with the EU’s horizontal framework under the EU AI Act (European Parliament and Council, 2024). The analysis identifies persistent challenges, including regulatory fragmentation, inconsistent application of principles and uneven enforcement capacities. It concludes that the UK model’s long-term effectiveness depends on stronger central coordination, proposing minimum statutory duties for regulators, a lead coordinating authority and clearer accountability pathways across sectors.