Welcome to my homepage!
I am a research professor at Simula Research Laboratory in Oslo, Norway, where I am Head of AI Security at the Center for AI Security and Safety (CAISS) and Head of the Department of Secure and Trustworthy Intelligent Systems (SENTRY). Our research investigates the interplay between artificial intelligence and software/system engineering, with a focus on robustness and security. Software systems increasingly combine conventional code with machine-learned components and AI agents. This brings new capabilities, but also new failure modes and attack surfaces. With my team, I develop data-driven techniques, tools, and agents that help keep such systems secure, resilient, and under control. This includes adversarial evaluation and red-teaming of large language models (LLMs) and AI-based systems, assessment of risks posed by AI in high-stakes and regulated contexts, and alignment of AI system design with emerging governance frameworks. Our work combines several fields, such as statistical program analysis, machine learning and AI, software repository mining, cybersecurity, and empirical software engineering.