A new artificial intelligence weather forecasting model developed by Google DeepMind has demonstrated superior accuracy and speed compared to the world’s leading conventional system, according to a study published in the journal Science. The model, named GraphCast, uses machine learning to analyze decades of historical weather data and current atmospheric conditions to predict future states up to 10 days in advance. In a head-to-head evaluation, GraphCast provided more accurate forecasts for over 90% of 1,380 test variables, including critical predictions for extreme weather events like tropical cyclones and atmospheric rivers. The AI system generates a 10-day forecast in under a minute on a single machine, a task that takes traditional numerical weather prediction models hours to compute on vast supercomputers. Researchers emphasize that AI models like GraphCast are not replacements but powerful complementary tools that can enhance the speed and resolution of existing forecasting infrastructure. The full study detailing the methodology and results is available in Science.