A novel generative AI model trained on data from the UK Biobank can forecast individual risk and timing of developing 1,258 diseases up to 20 years in advance. By integrating medical history with demographic and lifestyle factors, the model offers a powerful tool for early identification of high-risk patients across a broad disease spectrum. The approach adapts large language model architectures to medical data, providing precise and transparent multi-disease risk predictions that could revolutionize preventative medicine.