Society is entering a new era of catastrophes in which natural hazards are causing more damage than in the past. Recent years have seen a steep rise in economic and insured losses from weather and climate related hazards, largely due to a significant increase in exposure, but the likely scenario of the hazards themselves becoming more damaging in the future will only add to increased societal vulnerability. Recent failures in current risk management strategies have highlighted the need for a step change in our understanding of weather and climate risk. Earth system models are poised to bring about this change by providing new and independent risk assessments with respect to traditional practice based in historical observations. This talk will address the opportunities and challenges of incorporating dynamical models into climate risk assessment with a focus on tropical cyclone risk. Results will be presented from collaborations between NCAR and the reinsurance and energy industries to understand tropical cyclone risk and resulting impacts and losses. We are at the dawn of this new era of weather and climate risk assessment and these initial developments hold profound possibilities for the future. I will address the mathematical and computational challenges in achieving this step change in our understanding of climate risk.