AbstractThe quantification of damage potential is an essential prerequisite for mitigating the impact of natural hazards on the built environment of a country. Catastrophe models are complex computer programs that allow us to measure damage potentials, investigate the possible effects of climate change, and more. Model-generated information, however, is highly sensitive to the incorporated assumptions, and thus the soundness of our knowledge of risk and the effectiveness of mitigation mechanisms is in direct proportion to how well we understand those assumptions. The history of the development of catastrophe modeling in the last 60 years has much to teach us about these and other important aspects. However, it has been underexamined. This paper aims to trace the origins of modern catastrophe modeling—identifying its roots and rapid progress from World War II until the mid-1960s, when risk assessment approaches grew from empirical data-dependent techniques to physics-based computer simulations, and from the 1960s to early 2000s, when catastrophe models acquired an important role in policy analysis. The role of Don G. Friedman, a scientist not yet properly recognized in the risk modeling community although arguably the chief pioneer of catastrophe modeling, is highlighted. The sources used to construct this account include relatively unexplored ones as well as interviews with key pioneers of catastrophe modeling.