Appraising the Institutional Framework of the EU’s Artificial Intelligence Act (AI Act)
摘要
This chapter critiques the institutional framework of the European Union’s Artificial Intelligence Act (AI Act), the first comprehensive law to regulate AI globally. While the Act adopts a product safety model to leverage existing EU regulatory structures, it risks inheriting their limitations, raising questions about its ability to address AI’s unique challenges. The analysis identifies several concerns, including overlapping roles and unclear mandates among key institutions, such as the European Commission and the newly established AI Office. These inefficiencies are compounded by the Act’s integration into a broader, complex web of EU regulations, such as data and sector-specific laws, creating enforcement ambiguities and potential gaps in oversight. The chapter concludes by recommending clearer institutional delineations, streamlined governance mechanisms, and targeted reforms to align regulatory structures with the Act’s goals. By addressing these issues, the EU can strengthen its capacity to ensure effective AI governance.