Searching for Harmonised Rules: Understanding the Paradigms, Provisions and Pressing Issues in the Final EU AI Act

Abstract

This analysis provides an overview of the enactment of the final European regulation about harmonised rules on Artificial Intelligence (AI Act). The AI Act establishes the first legally binding horizontal regulation on AI. Our paper follows an interdisciplinary approach in combining legal scrutiny with political analysis in order to lay open the rationale, overall structure and shortcomings of the provisions of the new regulation. We understand the crafting of the AI Act as a reaction to the growing centralisation and power of non-European platforms in developing and providing AI systems, and EU's geopolitical and normative aspirations to shape the uptake of this technology. Overall, our contribution aims to familiarise researchers from other disciplines from tech to policy, humanities and social science with the complex regulatory structure and logic of the AI Act. We structure our analysis in three major parts: First, we analyse the regulatory necessity in introducing a coercive regulatory framework, and second, present the regulatory concept of the AI Act with its fundamental decisions, core provisions and risk typology. Lastly, a critical analysis points to shortcomings, tensions and watered down assessments of the Act.

Keywords: AI Act, AI Regulation, Standardisation, Product Safety Law, Geopilitics, EU Legislation, Fundamental Rights, Fundamental Rights Impact Assessment, Risk Regulation, legal definitions, AI, Governance, Ethical Guidelines, Digital Strategy