A new study demonstrates that artificial intelligence systems can generate highly accurate weather forecasts faster and with less computational power than conventional physics-based models. The research, published in a leading scientific journal, shows that machine learning algorithms trained on decades of historical weather data can predict key atmospheric variables like temperature and pressure with comparable or superior accuracy to established numerical models. These AI systems produce forecasts in minutes rather than hours, potentially enabling more frequent updates and earlier warnings for severe weather events. While promising, researchers note that further testing is needed on extreme events and long-term climate projections before operational deployment. For the full details, read the complete article at https://sciencedaily.com/releases/2024/07/240715140810.htm.