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 server, a task that typically requires hours on supercomputers using traditional numerical weather prediction methods. Initial testing shows the AI system matches or exceeds the accuracy of leading global forecasting models for predictions up to seven days ahead, particularly in tracking extreme weather events like tropical cyclones. While promising, researchers note the AI model currently lacks the granular detail of traditional systems for localized forecasts and requires further testing for operational reliability. The development signals a potential shift toward hybrid forecasting approaches that combine AI speed with physics-based model precision. Read the full article at https://technologyreview.com/2024/05/15/ai-weather-model-forecast-accuracy/