In 2026, we we will see the beginning of precise medical forecasting. Just as there has been remarkable progress in weather forecasting using huge language models, so too will there be in determining individual risk for major age-related diseases (cancer, cardiovascular and neurodegenerative diseases). These diseases share common features, such as a long incubation phase before any symptoms appear, usually two decades or longer. They also share the same biological underpinnings of immunosenescence and inflammation, terms that characterize an immune system that has lost some of its functionality and protective power, and the increased inflammation that accompanies it.
The science of aging has given us modern ways to track these processes using whole-body and organ clocks, as well as specific protein biomarkers. This allows us to determine whether a person or an organ within a person is aging at an accelerated rate. In addition, modern AI algorithms can see things that medical experts cannot, such as accurately interpreting medical images such as retinal scans to predict cardiovascular and neurodegenerative diseases years in advance.
These added layers of data can be linked to an individual’s electronic health record, which includes their structured and unstructured notes, lab results, scans, genetic results, wearable sensors, and environmental data. Together, this provides an unprecedented depth of information about an individual’s health, enabling risk prediction for three major diseases. Unlike polygenic risk scoring, which detects the risk of heart disease, common cancers and Alzheimer’s disease, precision medical forecasting takes it to a modern level by providing a predicted arc of time – the “when” factor. Analyzing all the data using huge reasoning models can reveal a person’s vulnerabilities and develop a personalized, aggressive prevention program.
We already know that the risk of these three diseases can be reduced with lifestyle factors such as an optimal anti-inflammatory diet, constant exercise and regular, high-quality sleep. However, in addition to addressing these factors, which are more likely to be implemented when a person is aware of their risks, we will have medications that will promote a fit, protective immune system and reduce inflammation throughout the body and brain. GLP-1 drugs have already been shown to be at the forefront of achieving these goals, but many more drugs are in the pipeline.
The potential for true medical prediction must be demonstrated and validated through prospective clinical trials that demonstrate, using the same markers of aging, that an individual’s risk is reduced. An example for people at increased risk of Alzheimer’s disease is a blood test called p-tau217and this risk can be significantly reduced by improving lifestyle, especially exercise. This can be confirmed using the brain organ clock and whole-body aging clocks.
This is a modern frontier in medicine – the potential for primary prevention of the three major age-related diseases that threaten our health and quality of life. This wouldn’t be possible without advances in both the science of aging and artificial intelligence. For me, this is the most invigorating future application of artificial intelligence in medicine: an unparalleled opportunity to prevent major diseases from occurring, something I dreamed of but was not possible at scale due to a lack of data and analysis. In 2026 it will finally be possible.