International
The Morning Science

What science found today

‹
 

Research

Putting Google's AI hurricane model to the test

Phys.org ·

Putting Google's AI hurricane model to the test

The Brief

  • A week before Hurricane Melissa hit Jamaica, forecast models disagreed on the storm's path and how strong it would become.
  • An AI model from Google DeepMind correctly predicted both the storm's path and its Category 5 strength.
  • The article compares this AI model's forecast with conventional forecasts for the same storm.

Why it matters

It offers a documented case where an AI hurricane model outperformed conventional forecasts on both track and intensity, which bears on how forecasters use such tools. A single storm does not establish overall accuracy.

Who’s affected

  • meteorologists and hurricane forecasters
  • people living in hurricane-prone coastal regions

A week before Hurricane Melissa slammed into Jamaica last fall, the forecast models couldn't agree on where it was headed or how strong it would get. One got it right—an AI model from Google DeepMind, which called both the storm's path and its Category 5 strength.

How do you feel about this?

0 Comments

    Putting Google's AI hurricane model to the test – The Morning Science