A new artificial intelligence weather forecasting model developed by researchers has demonstrated superior accuracy compared to conventional physics-based systems in global medium-range weather predictions. The AI model, trained on decades of historical weather data, can generate 10-day forecasts in minutes on a single computer, while traditional supercomputer-based systems require hours of computation. In head-to-head comparisons over a year of global weather, the AI system more accurately predicted key atmospheric variables like pressure, temperature, and humidity at various altitudes. The technology shows particular promise for improving predictions of extreme weather events, though meteorologists note that integrating AI with established physics models will likely provide the most reliable forecasts. The development marks a significant shift toward data-driven approaches in numerical weather prediction. Read the full article at: https://sciencedaily.com/releases/2024/05/240507150543.htm