Human oversight under the AI Act and its interplay with Art. 22 GDPR

The AI Act identifies human oversight as a fundamental mechanism for human control over AI systems to protect health, safety, and fundamental rights. While this may be promising from a policy perspective, there are some challenges to achieving effective hu-man oversight. This Article sheds some light on how the concept of human oversight might fit into the AI Act's overall regulatory approach to human-centric AI. Based on the concept of human oversight outlined by the AI High-Level Research Group, the Article analyzes the classification of human oversight measures under the AI Act, as well as the obligations and challenges arising from these provisions. In this context, particular attention is paid to the right under Art. 22 GDPR. Despite the challenges of the concept of human oversight, the different approaches under the AI Act and Art. 22 GDPR may work well together. Nevertheless, the effectiveness of human oversight depends in particular on effective tools in practice, and it remains to be seen whether the solution here lies in technology.     «

The AI Act identifies human oversight as a fundamental mechanism for human control over AI systems to protect health, safety, and fundamental rights. While this may be promising from a policy perspective, there are some challenges to achieving effective hu-man oversight. This Article sheds some light on how the concept of human oversight might fit into the AI Act's overall regulatory approach to human-centric AI. Based on the concept of human oversight outlined by the AI...     »