AI Can Forecast Your Health a Decade in Advance
Imagine if doctors could forecast your health like meteorologists predict the weather. Instead of “a 70% chance of rain tomorrow,” it might be “a 70% chance of developing type 2 diabetes in the next 10 years.” That’s no longer science fiction—it’s becoming reality.
Scientists have developed an artificial intelligence model, Delphi‑2M, that predicts a person’s likelihood of developing more than 1,200 diseases years in advance. Trained on millions of anonymised medical records, Delphi‑2M looks for patterns in a person’s health data—similar to how AI chatbots predict the next word in a sentence.
How It Works
- Delphi‑2M calculates probabilities of disease over time, not exact dates.
- Tested on UK Biobank data (~400,000 people) and validated on Danish medical records (~1.9 million people).
- Works best for predictable conditions such as type 2 diabetes, heart attacks, and sepsis.
The Big Question: How Will Health Systems Cope?
Forecasting risk for thousands of diseases could transform preventive medicine — but it raises critical challenges:
- Capacity: Can primary care and hospitals manage the follow-up for millions flagged as high-risk?
- Equity: Will predictive AI be accessible to all, or deepen existing inequalities?
- Integration: How will predictive insights feed into electronic health records, care pathways, and funding models?
- Trust: Patients need to believe the system is using their data responsibly and translating predictions into meaningful care. “AI can help us see health problems before they arrive, but unless health systems are prepared to respond, it risks overwhelming rather than empowering them. The challenge is not just prediction — it’s creating the infrastructure to act.” — Gareth Presch, CEO, WHIS
Investing in Prevention Like AI
Delphi‑2M shows what AI can do when we invest in prediction. Now imagine investing in prevention at the same scale: early detection programs, lifestyle interventions, and community health initiatives.
- Jobs: Expanding preventive care could create thousands of new roles across healthcare, digital health, and wellness sectors.
- Value: Reducing disease incidence lowers hospital costs, boosts productivity, and strengthens communities.
- Innovation: Preventive programs drive research, technology adoption, and new care models. “If we invested in prevention with the same ambition we give AI, we could not only prevent millions of illnesses but also create high-value jobs and deliver enormous economic and social returns.” — Gareth Presch
WHIS Perspective
At the World Health Innovation Summit, we see Delphi‑2M as part of a broader shift towards prevention, precision health, and community wellbeing. Predictive AI is exciting but the true potential lies in pairing forecasts with actionable care and system readiness.
References
- BBC News: AI can predict health problems over a decade into the future, say scientists (2025)
- The Guardian: New AI model can predict a person’s risk of more than 1,000 diseases (2025)
- Financial Times: New AI model predicts susceptibility to over 1,000 diseases (2025)
- medRxiv: Learning the natural history of human disease with generative transformers (2024)