Google unveils model with superior weather forecasting capabilities

Google Deepmind, an artificial intelligence company, has just announced a new artificial intelligence (AI) weather forecasting model that can provide 15-day forecasts with outstanding accuracy and speed. This is considered a potential tool to save lives in the context of increasing climate change.

The GenCast model is more accurate than the world's leading existing models.

According to the London-based company , the model, called GenCast, has proven to be more accurate than the world's leading models. Currently, the European Center for Meteorological Forecasting (ECMWF) provides forecasts for 35 countries and is considered the global benchmark for accuracy in the meteorological industry.

However, DeepMind representatives said that GenCast's forecasts outperformed ECMWF's accuracy in more than 97% of the 1,320 real-world scenarios tested in both from 2019. The new model was trained on 40 years of temperature, wind, and air pressure data from 1979 to 2018 and can produce a 15-day weather forecast in just eight minutes, much less than the current hourly rate.

DeepMind claims that GenCast provides more accurate forecasts for both daily weather and extreme events than leading forecasting systems . With its superior forecasting capabilities, DeepMind believes that this intelligent model can help limit the loss of life and property caused by natural disasters.

Scientists warn that extreme weather is becoming more common and severe due to climate change . In August 2023, a series of wildfires in Hawaii (USA) killed about 100 people. Locals criticized authorities for not being warned about the risk of wildfires early. This summer, an unusual heat wave in Morocco killed at least 21 people in 24 hours. Last September, Hurricane Helene claimed the lives of 237 people in Florida and other states in the Southeast of the US.