By: Justin Ding

Days before Hurricane Beryl hit Texas, an artificial intelligence software outperformed the current gold standard in weather forecasting

Hurricane Beryl was predicted to make landfall in Mexico’s Yucatan Peninsula as early as July 3, 2024, and stop, according to current American forecast models. Five days later, it had done what the models did not predict: it hit Texas and took at least 36 lives in the Houston area.

However, an AI created by Google Deepmind did predict that it would hit Texas in nearly the same spot and did it in less time. These accurate predictions could save more lives in a future with more and more natural disasters and do it very efficiently.

“It is clear why DeepMind’s accomplishment is deemed more heroic than even this: as extreme weather events become more commonplace globally, forecasters are in a race against time to develop software that can accurately predict catastrophes before they happen and help governments to evacuate people from danger zones,” reported Lauren Shirreff, who writes for The Telegraph.

As great as this AI, called Graphcast, is, it gets even better.
“GraphCast predicts weather conditions up to 10 days in advance more accurately and much faster than the industry gold-standard weather simulation system,” said Remi Lam, who graduated from MIT and is now a Staff Research Scientist at Google DeepMind.

In comprehensive tests, GraphCast outperformed HRES, from the European Centre for Medium-Range Weather Forecasts (ECMWF) in over 90% of evaluated weather variables.

In conclusion, this new AI called Graphcast created by Google Deepmind, is considerably better than traditional weather forecasting systems and is faster and more efficient. AI in the future can get more accurate weather forecasts and can help save more lives in natural disasters.