Projekt
Menschliche Aufsicht im europäischen Gesetz über künstliche Intelligenz und die Symbolik des Hammers
This thesis investigates the human oversight mechanism of the European Artificial Intelligence Act (AIA). An interdisciplinary approach was chosen to explore this normative safety mechanism, not only as a legal requirement, but as a symbolic affirmation of human agency in the digital age. It starts with a legal analys…
This thesis investigates the human oversight mechanism of the European Artificial Intelligence Act (AIA). An interdisciplinary approach was chosen to explore this normative safety mechanism, not only as a legal requirement, but as a symbolic affirmation of human agency in the digital age. It starts with a legal analysis of the relevant provisions of the AIA. A particular focus is placed on the obligations of providers and deployers of high-risk AI systems in accordance with Articles 13, 14, and 26 AIA. The fundamental practical difficulties in execution of this Article are highlighted, which pose a danger that human oversight will become a compliance mechanism rather than an effective safety mechanism. Furthermore, the right to explanation according to Article 86 AIA will be explained, which can be asserted by individuals whose rights are affected by decisions made by human overseers. Building on this legal basis, the work jumps into the methodology of cultural legal studies to examine the sensitivity and physicality of law beyond its linguistic dimension. This symbolic dimension is conceived through an artistic lens that uses the metaphor of the hammer, thereby expanding the legal interpretation. This applied symbolism was extracted from a personal artistic process. Therefore, the hammer is symbolic for the human authority and responsibility inherent in the obligation of human oversight. The hammer symbolizes the moment when human action asserts itself over automated decisions made by algorithms. Human oversight in the Artificial Intelligence Act is therefore developing not only as a regulatory requirement, but rather as a strong symbolic affirmation of the continuing primacy of human authority over AI systems. In the chosen transdisciplinary approach, it became clear that effective human oversight requires broad democratic and cultural engagement to ensure that AI governance remains human-centered and that this responsibility remains tangible in an era of increasing automated decision-making practices.