I am an Associate Professor of Cybersecurity in the Digital Security Group at the University of Oslo’s Department of Informatics, where I lead research on trustworthy AI-enabled and interoperable cyber defence.
My research examines how heterogeneous security observations, information, and domain knowledge can be transformed into structured, interoperable representations that support evidence-grounded reasoning, defensible decision-making, and controlled defensive action. I combine semantic modelling, ontologies, and knowledge graphs with symbolic and neuro-symbolic reasoning, machine learning, large language models, and constrained agentic systems. My central objective is to increase the analytical and operational capacity of cyber defence towards trustworthy autonomy by establishing the evidence, provenance, interoperability, security, accountability, and human and organisational control required for trustworthy AI-enabled cyber defence.
I initiate and lead competitively funded national and European research, translating scientific methods into experimentally evaluated prototypes, international cybersecurity standards, and open-source software. I teach master’s-level cybersecurity courses and supervise MSc and PhD research. I serve on the European Union Agency for Cybersecurity (ENISA) Advisory Group and the OASIS Open Board of Directors, and co-chair the OASIS CACAO Technical Committee.
Research Interests:
- Cyber knowledge and evidence-grounded reasoning: I study how security observations, threat intelligence, and contextual knowledge can be represented, integrated, and reasoned over to support threat-informed analysis and decision-making. My work combines semantic models and hybrid AI, with an emphasis on evidential support, provenance, and uncertainty.
- Interoperable security automation and defensive action: I investigate how cyber intelligence and defensive expertise can be translated into machine-readable playbooks and coordinated action across heterogeneous systems. My work addresses standards-based interoperability, AI-assisted response planning, and the validation and execution of defensive procedures.
- Trustworthy and governed autonomy: I study agentic systems for cyber investigation, planning, and response under explicit technical and organisational controls. My research examines how evidence quality, operational context, and expected consequences should determine the level of permitted autonomy, with operational authority established through validation, policy enforcement, and human authorisation.
- Security and assurance of AI-enabled cyber defence: I investigate attacks against AI-enabled defensive workflows and mechanisms for integrity protection, least-privilege execution, and secure tool interaction. Evaluation addresses robustness, traceability, policy compliance, and operational consequences as systems acquire greater autonomy.
I welcome enquiries from MSc and PhD students interested in exploring these topics, as well as from external researchers interested in collaboration. Please contact me to discuss potential research opportunities.
→ Visit my Google Scholar profile: [link]
→ Connect with me on LinkedIn: [link]