Can AI predict human lifespan ?

The ability of AI to predict human lifespan is a topic of interest in the field of medicine and healthcare. This would involve analyzing a wide range of factors, including genetic data, medical history, and lifestyle, to estimate an individual's likelihood of survival to a certain age. The prediction of human lifespan would have significant implications for healthcare, insurance, and social security. It would require a deep understanding of the complex interactions between genetic and environmental factors that influence human health and longevity. The potential applications of such a capability are vast, ranging from personalized medicine to pension planning. However, it also raises important questions about the potential impact on individual privacy and the role of AI in making decisions about human life.

Current AI systems can estimate remaining lifespan by combining electronic health records, lab values, imaging and sometimes lifestyle data; typical top-performing models (e.g., deep neural networks and gradient-boosted trees) achieve modest discriminative performance (concordance indices around 0.7–0.8 for large health-system cohorts) rather than precise individual predictions. Because lifespan is influenced by future accidents, evolving treatments and social determinants, prediction errors often widen for horizons beyond five years. Work continues on integrating polygenic risk scores and wearable streams, but regulatory-grade “AI fortune-telling” does not yet exist.

— Enriched May 11, 2026 · Source: The Lancet Digital Health — https://www.thelancet.com/journals/landig/article/PIIS2589-7500(21)00188-6/fulltext