In the article «The Economic Impact of Extreme AI Scenarios,» published in the North American Actuarial Journal, Prof. Dr. Martin Eling examines seven plausible, potentially high-consequence AI risk scenarios. Using the dynamic inoperability input–output model, the study considers scenarios including failures of widely used business software, breakdowns in AI-enabled transport systems, and targeted AI-based attacks on critical infrastructure.
The estimated losses for the U.S. economy range from US$11 billion to US$85 billion. While most scenarios fall within insurable limits, some could exceed the risk-bearing capacity of private insurers and require public–private risk-sharing mechanisms, particularly in the case of AI-based attacks on critical infrastructure.
Read the full article to explore the seven scenarios and what they reveal about the insurability of emerging AI risks.