Bettina Könighofer

Bettina Könighofer is an Assistant Professor at TU Graz’s Institute of Information Security and leads the Trusted AI Group. Her research focuses on trustworthy AI, formal verification, and safe reinforcement learning. She develops methods that provide formal safety guarantees for AI and autonomous systems, combining tools from reactive synthesis, runtime monitoring, and probabilistic verification.

Research Interests

• Runtime monitoring and enforcement for AI systems
• Safe reinforcement learning and shielding
• Reactive synthesis from temporal logic specifications
• Formal specifications for AI alignment
• Probabilistic verification and analysis