(Vienna, 08 August 2025) The Medical University of Vienna is represented on the international stage by an expert: Matthias Samwald from the Institute of Artificial Intelligence played a key role in drafting the new EU Code of Practice for General-Purpose AI (GPAI). This set of rules is intended to ensure that new AI models are developed under strict safety and quality standards. Leading AI developers such as OpenAI, Google, Anthropic, and xAI will base their work on it.
The Code of Practice, which was presented in Brussels, contains binding guidelines for dealing with the most powerful AI systems – from minimizing potential risks to promoting transparency and accountability. Samwald took the lead in the area of “Safety & Security,” which defines how risks such as misuse, lack of controllability, or systemic vulnerabilities can be identified and mitigated at an early stage.
“These rules have a direct impact in Europe, but are also expected to form the basis for international agreements and best practices,” emphasizes Samwald. “This creates the necessary foundation of trust. The task now is to use this foundation to massively accelerate the development of safe frontier AI and AI-driven innovation in Europe. That is my next challenge.”
The successful completion of the code marks an important step toward the safe development of future AI systems. The fact that an Austrian AI expert is at the forefront of these efforts underscores the international significance and expertise of MedUni Vienna in the field of artificial intelligence and its safe use.
Link: Matthias Samwald helps shape the world's first practice guidelines for artificial intelligence