On October 11, 2024, the joint event organized by CNDS and Climes explored the latest advancements in AI-driven weather forecasting, with a special focus on the detection and predictability of extreme weather events. Experts and attendees delved into the strengths and limitations of current AI models, as well as the broader social and ethical implications of shifting towards AI-based forecasting.

Leonardo Olivetti, PhD student at the Department of Earth Sciences, UU, kicked off the event with an overview of recent progress in deep learning models for weather prediction, highlighting key findings from his research. His presentation emphasized the remarkable achievements of deep learning in forecasting, while also acknowledging the areas where data-driven models still fall short compared to traditional physical models.
Emma Holmberg, PhD student at the Department of Earth Sciences, UU followed up with an informal opposition, adding nuance to the discussion by addressing some of the key shortcomings of deep learning models from a physical point of view, and their wider societal impact.

Expert Talks:
The seminar featured talks from several leading experts in the field:

• Clare Monteleoni, Professor of Computer Science at the University of Colorado Boulder and Research Director at INRIA, France, shared her expertise in machine learning, geospatial data, and climate modelling.
• Mengqian Lu, Associate Professor at the Department of Civil and Environmental Engineering, Hong Kong University of Science and Technology, discussed specific applications of AI in addressing hydroclimatic extremes, with a focus on atmospheric rivers.
• Ilaria Luise, Senior Research Fellow at CERN, Geneva, presented cutting-edge research on large-scale foundation models in the context of AI.
• Swetha Kolluri, World Bank, AI & digital innovations, who joined via Zoom shared her expertise on broader social and ethical implications of AI-based technologies.

The event concluded with a panel discussion moderated by Gabriele Messori, Professor of Meteorology at Uppsala University and Director of climes. The panellists engaged with the audience, addressing questions and exploring the latest trends in AI applications for Earth sciences and weather forecasting. They also reflected on the broader societal implications of these advancements, highlighting both the potential benefits and challenges of integrating AI into climate science.