The AI uses data from 199 earthquakes and the more than 130,000 aftershocks. Manual predictions are possible but they consume time and have proven to be less accurate than those of the AI.

Many believe aftershocks occur within a short timeframe of an earthquake. The AI, however, can predict aftershocks up to a year after the initial event.

No seismologists took part in the research, though two of its lead researchers – Phoebe DeVries of the Department of Earth and Planetary Sciences at Harvard, and Google AI recruiting lead Brendan Meade – consider themselves to be computational earth scientists.

Two other machine learning researchers from Google were involved with the project; Martin Wattenberg, and Fernanda Viégas.

The results of the project were published in the science journal ‘Nature’ today.

Nature came under fire earlier this year for a decision to lock its upcoming machine learning section behind a paywall.

Thousands of researchers are boycotting the journal as a result of their belief in open AI research. The researchers believe in sharing findings to both advance developments and promote ethical standards.

As of writing, the petition against the move features 3352 signatories.

What are your thoughts on the use of AI to predict natural disasters?

The post Harvard and Google partner on earthquake aftershock-predicting AI appeared first on AI News.