NASA’s COFFIES project used AI to improve space-weather prediction

NASA’s COFFIES project used AI to improve space-weather prediction. Researchers developed a machine-learning model capable of identifying regions likely to produce solar storms up to approximately 12 hours before they become directly visible. NASA

Earlier warning could protect critical systems. Solar activity can affect satellites, GPS, aviation communications, radio systems and terrestrial power grids.

This is a strong example of scientific AI complementing physical models. Machine learning detects patterns in large observational datasets, while domain models provide the scientific interpretation.

Source text: Raw news 8-20.docx — NASA.