The European Union´s Artificial Intelligence Act, approved on May 21, 2024, marks a significant regulatory milestone for businesses operating with artificial intelligence within the EU. One year after its adoption, HEC Paris Professor Pablo Baquero offered a comprehensive analysis of its impact, highlighting the Act´s innovative risk-based structure and its dual aims of promoting innovation and safeguarding fundamental rights. The Act addresses the growing need for harmonized rules across member states to support the safe circulation of products and services dependent on artificial intelligence while shielding citizens from technological misuse.
The legislation classifies artificial intelligence systems into four risk categories, each imposing a different level of regulatory obligation. Prohibited practices make up the highest risk tier, targeting applications such as predictive policing and real-time facial recognition in public spaces, which are outright banned. High-risk systems—common in sectors like education, recruitment, and public benefits—must adhere to rigorous data protection, transparency, explainability, non-discrimination, and cybersecurity standards before being introduced to the European market. Minimal-risk systems are required to maintain transparency, mainly by informing users they are interacting with artificial intelligence, while negligible-risk systems fall outside the Act´s scope, though industry best-practice guidelines may still apply.
For companies, the obligations are particularly pronounced for high-risk applications, with an emphasis on regular risk assessments and demonstrating compliance before market entry. Violations can result in severe financial penalties, the scale of which depends on the specific obligations breached. Notably, the Act´s rules are directed not only at artificial intelligence developers but also at those deploying or importing such technologies. While much of the technical detail remains under development, the current framework aims to balance the protection of individual rights against the need to foster innovation and economic growth. This nuanced approach positions the EU as a global regulatory leader in artificial intelligence, demonstrating the potential to shape both the trajectory of business innovation and the rights landscape across the continent.
