AI Foundation Model Aims to Make Stem Cell Therapies More Predictable

Researchers at the Blavatnik Institute at Harvard Medical School are contributing to the development of a new artificial-intelligence “foundation model” to improve the predictability and reproducibility of stem cell research, a major hurdle in regenerative medicine. By analyzing large biological datasets, the model seeks to identify the underlying rules governing cell differentiation, helping scientists anticipate outcomes and reduce reliance on trial-and-error approaches. The work has led to the formation of Cellular Intelligence, which was co-founded by researchers in the Blavatnik Institute at Harvard Medical School to translate these insights into practical tools for the field. Early work benefited from shared space and infrastructure at the Blavatnik Harvard Life Lab Longwood, illustrating how supportive translational environments can help bridge discovery science and real-world application. Read the article here.