The Brussels side-effect : how the AI Act can reduce the global reach of EU policy

Abstract

Over the last few years, artificial intelligence (AI) technologies have become embedded in various domains of social life, prompting legislative efforts at both national and international levels. In the European Union (EU), this drive for legislation has been reflected in various legal instruments, notably the proposed AI Act, which is expected to become a global standard through the “Brussels Effect.” This Article argues that while the AI Act will likely produce a Brussels Effect of its own, such an outcome will be accompanied by a side effect that undermines the EU’s ambition to spread legislative text and values in AI governance. Since the AI Act follows EU product safety legislation, its provisions supply limited protection to some of the values the EU policy intends to protect, such as the protection of fundamental rights. These shortcomings are compounded by the EU’s active efforts to shape alternative instruments, such as the Council of Europe’s proposed convention on AI along the lines of the AI Act. As a result, the diffusion of the AI Act as a global standard will have consequences for the EU policy agenda on AI and the conceptualization of the Brussels Effect.