AI tool forecasts 1,000 diseases decades ahead

A groundbreaking artificial intelligence (AI) tool is now capable of predicting a person’s risk of more than 1,000 diseases—sometimes up to 20 years before they occur.

Published Wednesday in the journal Nature, the study introduces Delphi-2M, a modified language model that analyzes health records and lifestyle factors to estimate the likelihood of conditions such as cancer, skin disorders, and immune diseases.

Unlike earlier AI models that focused on single illnesses, Delphi-2M can forecast the risk of 1,258 different diseases, offering doctors a powerful new way to spot high-risk patients early and take preventive action.

The tool draws on anonymized health data from 400,000 UK Biobank participants and 1.9 million patients in Denmark, and in most cases, it matched or outperformed existing disease-specific prediction models.

Experts say this breakthrough could transform healthcare by shifting the focus from treatment to prevention.