A new study from MIT’s Computer Science and Artificial Intelligence Laboratory (CSAIL) has developed a machine learning model that can significantly improve the accuracy of weather forecasting for extreme events like heatwaves and heavy rainfall. The model, called ClimaX, uses a novel ‘foundation model’ approach trained on diverse, global climate datasets to make predictions at different scales and for various atmospheric variables. Researchers report that ClimaX outperforms existing specialized models in predicting extreme weather patterns days in advance, offering a more unified and efficient tool for climate scientists. The advancement could lead to better early warning systems and more reliable long-term climate projections. Read the full article at: https://technologyreview.com/2024/07/15/climax-ai-weather-model.