Rosatom is applying artificial intelligence to speed up materials science for advanced nuclear reactors. According to TASS, AI cut the search for candidate alloy compositions from years to weeks, helping develop materials for the BREST fast neutron reactor and the VVER-SKD supercritical water-cooled reactor. The team screened around 250,000 alloying options and validated the chosen composition in the lab. AI is also being used to optimize carbon fiber production. The company says it reduced the scatter of strength characteristics along the fiber length and made carbon fiber structures lighter without losing strength. Rosatom plans to launch an automated synthesis complex in 2026 that closes the digital loop from AI-based property prediction to physical sample creation, which could cut material development time substantially. The work is part of Rosatom's broader push into AI, with collaboration involving the Kurchatov Institute and the Russian Academy of Sciences. The practical impact is still ahead, but the combination of AI and materials R&D could matter for both nuclear and non-nuclear low-carbon technologies.
