The National Oceanic and Atmospheric Administration (NOAA) has launched new AI-driven weather forecasting models designed to deliver faster and more accurate predictions while using far less computing power.

The AI models—AIGFS, AIGEFS, and Hybrid-GEFS—leverage decades of historical data and traditional equation-based models like the Global Forecast System to improve forecasts without replacing existing systems.

AIGFS can generate a 16-day forecast using just 0.3% of the computing resources of the traditional GFS, finishing in about 40 minutes, while AIGEFS provides a range of possible outcomes, and Hybrid-GEFS combines AI with traditional modeling to handle uncertainty.

NOAA estimates the AI programs require 91–99% less computing power than conventional models, though training them remains energy-intensive, CBS News has reported.

While promising, the agency notes further refinement is needed, particularly for hurricane forecasting and improving variability in ensemble predictions.