ETH Zurich AI Trains on NASA Climate Data
Analysis based on 6 articles · First reported Sep 11, 2026 · Last updated Sep 14, 2026
The initiative is a scientific research milestone with no immediate effect on public equity or commodity markets, though it could gradually benefit companies in weather analytics, disaster-risk modeling, and insurance. Broader market sentiment is likely unaffected, as the development carries no near-term revenue or policy implications.
Researchers at ETH Zurich in Switzerland have copied roughly 100 petabytes of publicly available United States — NASA climate data, about six billion files, onto servers adjacent to the Alps supercomputer at the Swiss National Supercomputing Centre in Lugano. The transfer took approximately one year. The team, led by computational physics professor Thomas Schulthess and climate physics professor Reto Knutti, is now training AI models on this data to accelerate and expand weather and climate forecasting and to detect patterns that traditional methods miss. According to the researchers, AI-based statistical models can run a global multi-day forecast in about a minute, versus hours with conventional equation-based simulation, and are cheap to run, enabling frequent iterations and early warning systems. They cite the Swiss village of Blatten, reportedly evacuated a week before a May 2025 glacier collapse, and the August 26 glacial collapse on the Nepal-China border, where satellite data reportedly showed warning signs, as examples of the potential to save lives. The work is a scientific and technological development with no direct corporate earnings impact, though it may influence weather forecasting, disaster prevention, and insurance industries over time.
Set up alerts, explore entity relationships, search across thousands of events, and build custom intelligence feeds.
Open Dashboard