A new method using artificial intelligence (AI) to solve complex quantum mechanical calculations for designing key catalysts in hydrogen fuel cells has been proposed. The National Research Foundation of Korea announced on the 26th that a joint research team led by Professor Minho Seo of Pukyong National University and Principal Researcher Seungmok Choi of the Korea Institute of Materials Science (KIMS) developed an integrated platform that predicts and verifies the activity and durability of platinum–cobalt (PtCo) fuel cell catalysts using AI. The results were published on June 2 (local time) in the international journal Advanced Functional Materials. The research team developed a neural network potential (NNP) model in which AI learns quantum mechanical calculation data and rapidly predicts energy and forces between atoms. They trained the AI on a dataset of actual quantum mechanical calculations for platinum–cobalt alloy catalysts, then used the model to predict the structure and stability of platinum–cobalt nanoparticles up to 5 nanometers in size. The calculations showed that in a “highly ordered structure,” bonding between atoms is strong, reducing the likelihood of metal dissolution and making stable performance more probable under long-term operating conditions. Durability tests experimentally verified the trend of increased stability.