This topic will be split into two parts for your listening convenience in order to keep each episode less than two minutes long. Both episodes covering Google and GraphCast will be published at the same time. 
Yesterday, on the 14th of November, Google’s DeepMind published a peer-reviewed study in the journal Science introducing their GraphCast technology.  
GraphCast is a weather forecasting system that uses machine learning and Graph Neural Networks (GNNs). In other words, it’s an AI model for weather prediction.
This model is capable of delivering 10-day weather predictions with a high level of accuracy and is considerably faster than the current industry gold standard when it comes to weather simulation systems; the High Resolution Forecast (HRES), produced by the European Centre for Medium-Range Weather Forecasts (ECMWF). 
“While GraphCast’s training was computationally intensive, the resulting forecasting model is highly efficient.” wrote Remi Lam, a staff research scientist at Google DeepMind. “Making 10-day forecasts with GraphCast takes less than a minute on a single Google TPU v4 machine. For comparison, a 10-day forecast using a conventional approach, such as HRES, can take hours of computation in a supercomputer with hundreds of machines.”
As Melissa Heikkilä wrote for the MIT Technology Review, “GraphCast outperformed the model from the European Centre for Medium-Range Weather Forecasts (ECMWF) in more than 90% of over 1,300 test areas. And on predictions for Earth’s troposphere—the lowest part of the atmosphere, where most weather happens—GraphCast outperformed the ECMWF’s model on more than 99% of weather variables, such as rain and air temperature”.