AI and cyber insurance: What risks do we need to understand?

Karlsruhe, October 07, 2026 — Are risks posed by artificial intelligence fundamentally different from known cyber risks? And how can we determine what happened when damage occurs? These questions were the focus of a seminar held on October 2 at FIZ Karlsruhe’s Berlin office.

Colleagues from our Intellectual Property Rights unit, headed by Prof. Dr. Franziska Boehm, met there with Prof. Dr. Rainer Böhme from the University of Innsbruck. He is a professor of information security and data protection and conducts research on topics such as digital forensics and the economic aspects of cybersecurity, including cyber insurance. The seminar focused on current and future damage scenarios caused by AI and their implications for cyber insurance.

The discussion focused on legal and technical options for addressing these risks. Where do we need more research and development to better understand vulnerabilities in AI? What role does AI play in relation to vulnerabilities in conventional software?

Another key focus was “forensicability”: How must systems be designed so that the sequence of events and the causes of an incident can be reconstructed after the fact? This brought into focus not only risk management issues but also the question of how damage can be investigated in a traceable manner.