A Korean research team developed an AI model, KIST-Ocean, that learns 160-plus years of global ocean data and predicts ocean conditions for the next 200 days in a matter of seconds. Researchers Kang Dae-hyeon and Kim Jeong-hwan from the Korea Institute of Science and Technology (KIST) Climate and Carbon Cycle Research Center announced the development on the 26th, with findings published in the journal Science Advances. The model was pre-trained on global large-ensemble simulation data from 1850 to 2014 and fine-tuned with ocean reanalysis data from 1982 to 2013. KIST-Ocean uses 62 ocean variables to predict 3D ocean states down to a depth of 600 meters at five-day intervals, incorporating six surface boundary conditions. The team stated that KIST-Ocean completed pre-training in about 33 hours and fine-tuning in about 2.4 hours using one Nvidia A100 GPU, generating global ocean forecasts for 200 days in 6–7 seconds. In the 2015 super El Niño case, it successfully reproduced the rise in sea conditions.