Regulatory Architectures for Autonomous AI Agents: A Comparative Jurisprudential Analysis of UK and EU Frameworks

Abstract The proliferation of autonomous AI agents capable of independent economic decision-making has precipitated a paradigm shift in digital governance, necessitating novel regulatory responses. This paper examines the divergent legal frameworks emerging in the European Union and the United Kingdom to govern the "AI Agents Economy"—ecosystems in which algorithmic systems negotiate, contract, and act without direct human oversight. Through comparative doctrinal analysis, this study evaluates the EU's risk-based Artificial Intelligence Act (2024) against the UK's principle-led, sectoral approach, focusing on legal personhood, contractual validity, liability attribution, and algorithmic accountability. Findings reveal a fundamental tension between innovation imperatives and accountability mechanisms, with both jurisdictions rejecting electronic personhood while grappling with liability gaps in autonomous systems. The analysis proposes a hybrid "responsibility mapping" model integrating mandatory insurance schemes, algorithmic audit trails, and cross-border regulatory cooperation to address governance lacunae. This research contributes to emerging scholarship on machine agency by offering a jurisprudential framework for reconciling technological autonomy with democratic legitimacy in digital transformation contexts. Keywords: Autonomous AI Agents; AI Governance; EU Artificial Intelligence Act; Algorithmic Accountability; Comparative Digital Regulation; Liability Frameworks for AI