First Analysis of the EU Artifical Intelligence Act: Towards a Global Standard for Trustworthy AI?

The EU AI Act represents unprecedented global regulatory scope: As of August 1, 2024, the EU AI Act is the first binding, horizontal legislation that applies across entire AI value chains—affecting not just companies selling AI in Europe, but all organizations involved in developing and deploying these systems for the EU market, regardless of where they operate.

Compliance is phased in progressively: Rather than imposing all requirements at once, the Act staggers compliance obligations over time, allowing organizations in the global AI supply chain to adapt their practices incrementally rather than facing immediate, comprehensive regulatory demands.

The Act aims to set a global standard for trustworthy AI: The paper identifies the Act's "global appeal in support of trustworthy AI" as both a major strength and a significant challenge, suggesting that while the EU's approach could influence international AI governance, achieving widespread adoption of these standards globally remains uncertain.

The legislation applies to both private and public organizations: The Act's scope extends beyond commercial companies to include public sector organizations, meaning government agencies and institutions using AI systems in the EU must also comply with the same regulatory framework.

Global value chain accountability is a key feature: By imposing requirements on all participants in the AI development and deployment chain—not just final product sellers—the Act attempts to ensure responsibility and trustworthiness throughout the entire process of creating and using AI systems.