Researchers in Switzerland are using artificial intelligence (AI) and extensive climate data from NASA to improve the prediction of natural disasters, with the goal of saving lives through faster and more accurate forecasts. A team at the Swiss Federal Institute of Technology in Zurich (ETH Zurich) has transferred approximately 100 petabytes of publicly available NASA climate data to servers connected to one of the world's most powerful supercomputers, called Alps. To give a sense of scale, this data is equivalent to about 20 million feature-length movies and is being used to train AI models that can enhance weather and climate forecasting capabilities. Thomas Schulthess, a professor of computational physics at ETH and head of the Swiss National Supercomputing Centre (CSCS), described this effort as a major scientific advancement. He explained that having NASA’s climate data directly linked to the supercomputer opens up new possibilities for research that were previously unattainable. Reto Knutti, a climate physics professor and head of ETH’s Center for Climate Systems Modeling (C2SM), noted that the data transfer process took about a year and involved moving roughly 6 billion files. The AI models being developed from this data can generate weather forecasts much faster than traditional methods. Knutti described these new statistical models as "really, really powerful" and "incredibly fast," capable of running global weather forecasts for several days in just a minute. These models are also cost-effective to run, allowing scientists to perform frequent simulations to detect specific weather patterns. This capability is especially valuable for early warning systems that can alert communities to potential disasters. The importance of early detection was highlighted by past events. For instance, a Swiss village called Blatten was evacuated a week before a glacier collapsed in May last year, based on satellite data that had indicated the risk more than a year earlier. Similarly, satellite image analysis had identified warning signs before a major glacial collapse on the Nepal–China border in August, which could have been monitored more closely if the data had been analyzed in time. The researchers emphasized that data is central to their work. Knutti stated that "data is essentially everything," and the close integration of the data with the supercomputer's immense processing power is crucial. This setup allows for rapid analysis and processing, which is vital for delivering timely and accurate forecasts that can help prevent loss of life and property.