David Wood, Managing Director at JBA Risk Management, and Jochen Papenbrock, Head of Financial Technology (EMEA) at NVIDIA, joined Matthew Grant to discuss the evolving landscape of catastrophe modelling powered by Artificial Intelligence. The conversation centers on how accelerated GPU computing and generative modelling are advancing the accuracy and scale of weather and flood risk analysis.
The discussion explores JBA’s early move to GPU computing, which enabled large-scale, national flood mapping long before such approaches became industry standard. NVIDIA’s evolution from gaming graphics to pioneering physics-informed Artificial Intelligence models has been pivotal, allowing for high-resolution simulations that capture far more complex and diverse climate events. Ensemble modelling, now capable of running over a thousand simulations, empowers analysts to predict rare or extreme floods that previous historical records could not reveal.
Artificial Intelligence models, such as those deployed by JBA and NVIDIA, are enabling the creation of richer, more varied flood event sets for regions like Europe and beyond. These advances support both scenario planning and real-time event monitoring, making it possible to simulate how a flood or catastrophe might unfold as it occurs. Earth-2 Studio and open-source tools are making generative models easier to use for catastrophe modellers, democratizing access to cutting-edge risk analytics. The partnership between JBA and NVIDIA has demonstrated the tangible benefits of faster, more agile modelling, while the speakers emphasize that broad, cross-sector collaboration is essential to fully realize the transformative potential of Artificial Intelligence and accelerated computing in catastrophe risk management.
