Artificial intelligence (AI) has successfully outperformed conventional weather forecasting methods for the first time. GraphCast, a new AI model from Google DeepMind, is able to predict precise weather forecasts in less than a minute – all from a desktop computer. 

Traditional weather prediction methods, which typically rely on physics equations, are infamous for being extremely time-consuming and costly.

Google DeepMind claims that through deep learning and data, the time and cost are all but removed. 

GraphCast has learned about the cause-and-effect relationships of the Earth’s weather patterns through decades of historical data. DeepMind says this is what has allowed the model to produce better forecasts than traditional methods.

A peer-reviewed paper published in Science on Tuesday (14 November) found that AI was more accurate than the world’s most advanced forecasting system. 

The European Medium Range Weather Forecasting model was beaten on 90% of the 1,380 metrics tested, including wind speed, temperature and air pressure. 

How well do you really know your competitors?

Access the most comprehensive Company Profiles on the market, powered by GlobalData. Save hours of research. Gain competitive edge.

Company Profile – free sample

Thank you!

Your download email will arrive shortly

Not ready to buy yet? Download a free sample

We are confident about the unique quality of our Company Profiles. However, we want you to make the most beneficial decision for your business, so we offer a free sample that you can download by submitting the below form

By GlobalData
Visit our Privacy Policy for more information about our services, how we may use, process and share your personal data, including information of your rights in respect of your personal data and how you can unsubscribe from future marketing communications. Our services are intended for corporate subscribers and you warrant that the email address submitted is your corporate email address.

Rémi Lam, a staff research scientist at Google DeepMind who helped create the model, said: “The main advantage of this AI approach is that it’s extremely accurate.

“It learns from decades of data and is able to be more accurate than the industry gold standard.”

Through the use of data and deep learning, GraphCast can evolve to understand varying weather patterns in a changing climate, Deepmind said.

However, Lam believes that AI is not ready to completely overtake traditional methods yet.

“AI models are trained from data and that data is generated by traditional approaches, so we still need the traditional approach to gather data to train the model,” Lam said in an interview with the BBC.

Graphcast is open source which means anyone can use the technology.

Technology companies around the globe have been designing their own AI weather predicition tools, including EU-funded Copernicus Climate and Huawei’s Pangu-Weather.

In April 2023, Google’s parent company, Alphabet, announced that it was merging its internal AI research team, Google Brain, with DeepMind.

Google’s CEO, Sundar Pichai, said that the merger would bring “together two world-class AI teams” that “will accelerate our progress in AI and help us develop more capable AI systems more safely and responsibly.”