Thu May 22 04:30:23 UTC 2025: Here’s a summary of the article and a rewritten version as a news article:

**Summary:**

Microsoft has developed a groundbreaking AI model called Aurora that significantly outperforms traditional methods in weather forecasting, hurricane tracking, and air quality prediction. Trained solely on historical data, Aurora has proven more accurate, faster, and cheaper than established forecasting centers like the U.S. National Hurricane Center and even the highly regarded European Centre for Medium-Range Weather Forecasts (ECMWF). This development, following similar advancements from companies like Huawei and Google, signals a potential paradigm shift in weather forecasting, with major weather agencies now actively exploring AI learning models of their own.

**News Article:**

**AI Revolutionizes Weather Forecasting: Microsoft’s Aurora Outperforms Traditional Methods**

**Paris – May 22, 2025** – A new artificial intelligence (AI) model developed by Microsoft is poised to revolutionize weather forecasting, surpassing traditional methods in accuracy, speed, and cost-effectiveness. The system, dubbed “Aurora,” has demonstrated an unprecedented ability to predict weather patterns, track hurricanes, and monitor air quality with greater precision than established forecasting centers worldwide.

According to a study published in *Nature*, Aurora, trained exclusively on historical data, outperformed operational forecasting centers like the U.S. National Hurricane Center in predicting the paths of hurricanes in 2023. In one striking example, Aurora accurately predicted the landfall of Typhoon Doksuri in the Philippines four days in advance, a feat unmatched by official forecasts at the time.

“For the first time, an AI system can outperform all operational centers for hurricane forecasting,” said Paris Perdikaris, Associate Professor of Mechanical Engineering at the University of Pennsylvania and senior author of the study.

The model also surpassed the accuracy of the European Centre for Medium-Range Weather Forecasts (ECMWF), considered the global benchmark for weather prediction, in 92% of cases for 10-day global forecasts. Notably, Aurora achieves this with significantly lower computational costs – several hundred times less than traditional models.

This breakthrough follows similar advancements from other tech giants, including Huawei’s Pangu-Weather AI model, and Google’s GenCast, suggesting a potential paradigm shift in how weather agencies approach forecasting.

Florence Rabier, Director General of the ECMWF, acknowledged the potential of AI, stating that their organization has been taking the technology “very seriously.” Meteo-France is also developing its own AI learning model to complement traditional forecasting methods.

While Aurora remains an experimental system, its success underscores the transformative power of AI in predicting increasingly severe and unpredictable weather events driven by global warming. Experts believe that within the next decade, advanced AI systems will be able to directly integrate data from satellites and weather stations to generate high-resolution forecasts anywhere in the world.

Read More