Google DeepMind says its WeatherNext model can accurately predict a storm's track and intensity using lower-resolution weather data, and open sources the model (Victoria Turk/Wired)

DeepMind's WeatherNext model has made a groundbreaking advancement in cyclone forecasting, enabling predictions with an additional day of warning. This breakthrough, detailed in a Nature paper, promises to save countless lives and billions of dollars in damages every year. Tropical cyclones, also known as hurricanes or typhoons, have caused over 700,000 deaths and $1.4 trillion in economic losses globally over the past 50 years.

Forecasts are crucial for timely warnings, but current methods are limited, often providing only a two-day lead time. WeatherNext, an AI model developed by Google DeepMind and Google Research in collaboration with meteorological experts, has achieved state-of-the-art accuracy in predicting a cyclone's track, intensity, and wind structure.

By integrating global weather dynamics with historical cyclone observations, the model learns complex atmospheric patterns and extreme weather behavior. When presented with a 15-day forecast, WeatherNext generates up to 1,000 possible scenarios, covering rare but devastating events like rapid intensification, as seen during Hurricane Melissa in 2025.

This model operates at a coarser 28x28km resolution, 100 times less detailed than traditional systems, yet still produces highly accurate intensity forecasts. WeatherNext has already proven its value, assisting the National Hurricane Center in predicting Hurricane Melissa's rapid intensification and landfall in Jamaica, allowing for advance warnings and critical preparation.

To further advance the field, DeepMind is open sourcing WeatherNext 2 and WeatherNext Cyclones models, as well as a smaller, lighter version, WeatherNext 2-mini, for use by researchers, forecasters, and organizations worldwide. These tools aim to accelerate progress in weather prediction and disaster preparedness, ultimately protecting lives and infrastructure from extreme weather events.