Delphi-2M is the innovative artificial intelligence model developed by researchers from the European Bioinformatics Institute, the German Cancer Research Center (DKFZ), and the University of Copenhagen. This system stands out for its ability to predict the risk of developing over 1,000 diseases based on patients' medical history and personal data, marking a significant advance in preventive medicine. The study, published in the journal Nature, details how Delphi-2M employs an advanced architecture based on generative transformers, similar to those used in models like ChatGPT, but adapted to analyze the temporal progression of human diseases. Thanks to this technology, the model can anticipate the long-term evolution of pathologies, providing healthcare professionals with a more accurate view of the development and interaction of different diseases. For its training, Delphi-2M used data from 402,799 individuals from the UK Biobank, collected before July 1, 2020. It was subsequently validated with information from 1.93 million Danish citizens recorded between 1978 and 2018. This robust database has allowed the model to predict with high accuracy the incidence of diseases such as breast cancer, cardiovascular diseases, diabetes, asthma, depression, acute myocardial infarction, chickenpox, and even mortality, considering factors such as previous diagnoses, lifestyle habits (alcohol and tobacco consumption), age, sex, and body mass index. One of the most notable features of Delphi-2M is its ability to generate synthetic future health trajectories, allowing for the estimation of the potential disease burden up to 20 years ahead. Additionally, the model has identified patterns of comorbidity among different diseases and shed light on how medical history can influence the emergence of new pathologies. For example, it was observed that cancer consistently increases the risk of mortality, while the effects of a myocardial infarction or sepsis tend to diminish after five years. However, researchers emphasize the importance of interpreting the results of Delphi-2M with caution, as there may be biases arising from the training data. They also stress that this model does not replace professional medical care but should be considered a complementary tool to enhance decision-making in personalized health and precision medicine. Although Delphi-2M still requires further validation before clinical implementation, its analyses provide valuable projections about the future evolution of health, reflecting the influence of past medical events on the development of new diseases.