In my presentation, I introduced the EU AI Act as a key instrument for systematically addressing the risks of artificial intelligence. It became clear that AI poses not only technical challenges, but also profound social and ethical challenges — for example, through discrimination, lack of transparency, or interference with fundamental rights. The risk-based approach of the law distinguishes between minimal, limited, high, and unacceptable risk and sets different requirements depending on the classification. I particularly emphasized that AI is a socio-technical system in which humans are not only users but also affected parties — and that standards, transparency, and responsible data governance are crucial to ensuring trust and security.