(Image credit: Getty Images/da-kuk)
A new artificial intelligence (AI) model has the ability to predict key atmospheric events faster and with greater accuracy than some of the world's most popular forecasting systems.
The model, called Aurora, is trained on more than 1 million hours of global atmospheric data, including weather station data, satellite imagery, and radar data. Microsoft researchers say it’s likely the largest dataset ever used to train an AI weather model.
Aurora accurately predicted that Typhoon Doksuri would hit the northern Philippines four days before it made landfall in July 2023. While official forecasts predicted the storm would be over Taiwan, which was hundreds of miles off its true location.
You may like
-
AI can handle tasks twice as complex every few months
-
What is Artificial Superintelligence (ASI)?
-
AI Lies, Dripping North America, and More
Aurora also outperformed traditional forecasting tools used by agencies like the U.S. National Hurricane Center and Joint Typhoon Warning Center. It provided more accurate five-day storm tracks and generated high-resolution forecasts up to 5,000 times faster.
Sourse: www.livescience.com
