A new artificial intelligence model called Aurora has demonstrated the ability to generate accurate global weather forecasts in under a minute, significantly faster than traditional physics-based models. Developed by researchers at Microsoft, Aurora was trained on over a million hours of diverse weather and climate data. In initial tests, the model produced five-day global air pollution predictions and ten-day weather forecasts that matched or exceeded the accuracy of established systems like the European Centre for Medium-Range Weather Forecasts (ECMWF). The AI model’s speed and efficiency could enable more frequent updates and higher-resolution forecasts, potentially improving severe weather warnings. However, experts note that further rigorous testing on real-world, operational data is needed to fully validate its capabilities and reliability for critical decision-making. Read the full article at: https://technologyreview.com/2024/06/12/1093605/microsoft-aurora-ai-weather-forecasting-model/