Google Deepmind's New AI Model GenCast To Deliver Faster, More Accurate Weather Results

In a paper published in Nature, Deepmind says that the AI model GenCast outdo the European Centre for Medium-Range Weather Forecasts ENS, which is the top operational medium-range weather forecast in the world.

GenCast marks a critical advance in AI-based weather prediction that builds on our previous weather model, which was deterministic, and provided a single, best estimate of future weather. By contrast, a GenCast forecast comprises an ensemble of 50 or more predictions, each representing a possible weather trajectory, Deepmind said in a Blogpost.

Deepmind team said that they have trained the AI model with the weather data upto 2018. Compared to forecasts for 2019, GenCast was more accurate than ENS on 97.2% of the time.

The tech giant said that GenCast is part of their growing suite of next-generation AI-based weather models, starting to enhance user experiences on Google Search and Maps. Google added that they will soon release real-time and historical forecasts from GenCast and previous models, enabling anyone to use those weather inputs into their own models and research.