AI Regulation in Focus
Research at the Karlsruhe Institute of Technology (KIT) is contributing to one of the central debates of the digital age: How should artificial intelligence be regulated? For their work addressing this question, T.-T. Prof. Dr. iur. Frederike Zufall and Dr. iur. Gustavo Gil Gasiola, together with co-authors from the Max Planck Institute for Security and Privacy, have been awarded the CNIL/EHESS Award 2026.
The award was presented on June 24 at the Privacy Research Day, an annual event hosted by the French data protection authority CNIL. In 2026, the event took place in conjunction with the G7 meeting in Paris. The highly competitive and internationally recognized prize honors outstanding social science research in the fields of data protection, privacy, and digital society.
The award-winning paper, “Do Citizens Agree with the EU AI Act? Public Perspectives on Risk and Regulation of AI Systems,” examines the extent to which European AI regulation aligns with citizens’ expectations. It is based on a large-scale study involving more than 1,400 participants in Germany, France, Spain, and the United States, who evaluated 48 different AI applications. The study design operationalizes the regulatory concept of the EU AI Act by linking its risk-based approach to the potential impact of AI systems on fundamental rights and other legally protected values.
The findings reveal a clear pattern: regardless of the specific AI system, many people perceive AI as posing risks to their fundamental rights as well as to protected values such as the rule of law and democracy. Accordingly, a majority supports broad regulation of AI systems—not only those classified as high-risk—while at the same time favoring a more moderate regulatory approach overall. In this respect, the findings challenge key assumptions of the EU AI Act, which relies on a tiered system of regulation based on risk categories.
“The question of how AI should be regulated is not only a legal one, but also a technical one. With our study, we provide a systematic insight into how people evaluate different AI systems in light of their perceived relevance for fundamental rights, the rule of law, and democracy. As these findings suggest a tension with the current concept of the AI Act, it raises the question of whether future AI regulation should be based on other—potentially technical—characteristics rather than on a tiered risk-based framework,” says co-author Tenure-Track Professor Frederike Zufall, Chair of Public Law and Computer Science at the KIT Department of Informatics.
The CNIL/EHESS Award highlights the societal relevance of this research. At the same time, it demonstrates the important role of interdisciplinary and international collaboration in aligning technological innovation with societal expectations and values.