Since December 2025, Frank Köster has been the new Professor of AI Safety and Security at the Faculty of Computer Science. At the same time, he continues to serve as the founding director of the Institute for AI Safety and Security at DLR. With this joint appointment based on the Jülich model, DLR and the Faculty of Computer Science are growing closer together in research and teaching.

Artificial intelligence is now used in all areas of everyday life, including those with particularly high security requirements, such as critical infrastructure or the handling of confidential data.

Founded in 2020, the Institute for AI Safety and Security researches and develops AI-related methods, processes, algorithms, technologies, and system environments. AI safety and security is viewed from two perspectives: on the one hand, it is about protecting AI systems from external attacks (security), and on the other hand, it is about ensuring their accuracy and operational safety (safety). The focus here is on the fields of transportation, energy, aerospace, and Industry 4.0. The researchers’ goal is to advance the operational and defense security of AI-based solutions in ambitious application classes.

Frank Köster is particularly interested in the following topics:

• Industrialization of AI and AI-based solutions in safety-critical system contexts and application situations with high security requirements. A key aspect here is the ability to connect excellent research results to practical needs and existing and emerging norms/standards and regulatory frameworks for the use of AI.
• Establishment of AI engineering based on improved and novel AI methods and algorithms. Safety and security are integral aspects of engineering – “by design.”
• Systematic process and IT support for the entire AI lifecycle, including successive improvements in AI and learning.
• Reliable and trustworthy embedding of AI-based solutions in data-supplying system contexts (including decentralized data/service ecosystems).
• Ethical, legal, and social aspects in the context of AI in the sense of an interdisciplinary/transdisciplinary research approach

Frank Köster emphasizes: “I am looking forward to working with my colleagues at Ruhr University Bochum. The Bochum location enables excellent research in the important field of AI. Topics related to safety and security can be easily integrated into already established research activities at the university and associated research institutions.”