• Google DeepMind has unveiled GenCast, its AI tool for weather forecasting.
• According to Google DeepMind, GenCast outperformed existing forecasting systems in trials.
• Better forecasts would enable better preparation for extreme weather events, the company said.

Google DeepMind announced its new artificial intelligence weather forecasting tool called GenCast on Wednesday.

According to Google DeepMind, GenCast differs from other models because it “adapts to Earth’s spherical geometry and learns to accurately generate the complex probability distribution of future weather scenarios when given the most current weather state as input.”

As a result, Google DeepMind said GenCast delivers better forecasts than the “top operating system,” citing the European Center for Medium-Range Weather Forecasts and its model – known as ENS – which makes forecasts up to 15 days in advance. GenCast, trained on the European center’s data, “consistently outperformed ENS” when predicting extreme heat, extreme cold and high wind speeds.

“Now consider tropical cyclones, also known as hurricanes and typhoons. Getting better and more advanced warnings about where they will make landfall is invaluable. GenCast provides superior forecasts of the tracks of these deadly storms,” the company said.

In a paper published Wednesday in Nature, Google DeepMind researchers wrote that GenCast “had greater capabilities than ENS on 97.2% of the 1,320 targets we evaluated.”

“Better predictions of extreme weather conditions enable better decisions,” DeepMind said in its press release. The company also announced it would share real-time and historical forecasts from GenCast.

“We are committed to collaborating with the broader weather community, including academic researchers, meteorologists, data scientists, renewable energy companies and organizations focused on food security and disaster relief,” the company said.