What the software does
The software uses machine learning to tackle problems that have traditionally required weeks of simulations. In materials science, researchers often model atomic interactions to find new substances. The AI aims to cut that process from months to days by spotting patterns in the data. For dark matter, the challenge is different. Scientists sift through petabytes of detector and telescope data looking for faint signals. The new software is designed to analyze that data more efficiently, the company says.
Dark matter is thought to make up about 85% of the universe's mass, yet it has never been directly detected. Experiments rely on underground detectors, particle colliders, and space telescopes. The data from these instruments is massive and noisy. AI can help separate potential dark matter signals from background noise, a task that has become a bottleneck in the field.
Why ISC 2026
ISC is one of the largest annual gatherings for high-performance computing. Researchers from national labs, universities, and tech companies attend to see the latest hardware and software. By debuting the software there, NVIDIA is signaling that scientific computing remains a core focus. The company's GPUs already power many of the world's fastest supercomputers. The new AI layer is meant to make those systems smarter, not just faster.
Materials discovery is another area where computing has become essential. New materials drive advances in batteries, semiconductors, and medical implants. The traditional trial-and-error approach is slow and expensive. Computational screening can narrow down candidates, but it still
