New AI Tool , Outperforms Traditional , Forecasting Models. Researchers say that artificial intelligence could save lives by helping to alert people about hurricanes much faster than traditional forecasting methods. Researchers say that artificial intelligence could save lives by helping to alert people about hurricanes much faster than traditional forecasting methods. BBC reports that a new Google DeepMind tool, GraphCast, predicted that Hurricane Lee would make landfall in Canada three days faster than other forecasts. Scientists say that AI's ability to quickly use an analysis of past events to make accurate predictions is a game-changer. Scientists say that AI's ability to quickly use an analysis of past events to make accurate predictions is a game-changer. AI-driven alert systems may be able to provide communities critical time to prepare ahead of extreme weather events like storms, heat waves and floods. . AI-driven alert systems may be able to provide communities critical time to prepare ahead of extreme weather events like storms, heat waves and floods. . Hundreds of factors are used to make accurate weather predictions, including air pressure, wind speed, temperature and humidity. . According to a paper published by DeepMind in the journal 'Science,' GraphCast outperformed existing models on over 90% of those factors. BBC reports that GraphCast's approach also produces forecasts in under a minute while requiring just a fraction of the computing power used by traditional forecasting methods. GraphCast uses machine learning to sort through vast amounts of data in order to learn how weather patterns could evolve. GraphCast uses machine learning to sort through vast amounts of data in order to learn how weather patterns could evolve. The main advantage of this AI approach is that it's extremely accurate, Remy Lam, Google DeepMind, via BBC. It learns from decades of data and is able to be more accurate than the industry gold standard, Remy Lam, Google DeepMind, via BBC