Causal AI for real-world public health decisions

In June 2026, Cambridge Public Health hosted Professor Mihaela van der Schaar, Professor of Machine Learning, AI and Medicine at the University of Cambridge, for a seminar exploring the potential of causal AI to support decision-making in public health and healthcare.

The seminar examined how causal AI can move beyond prediction to estimate the likely effects of interventions, helping researchers and policymakers answer questions about what might happen under different scenarios. Drawing on more than a decade of work in machine learning and causal inference, Professor van der Schaar outlined how these approaches can provide more actionable evidence for complex public health challenges.

A major focus of the talk was the emerging concept of AI-enabled digital twins – virtual models that can simulate possible futures, test alternative interventions and support real-world decision-making. The seminar explored how digital twins could be used to evaluate policies, improve healthcare delivery and help public health systems make better-informed decisions in the face of uncertainty.

The session was chaired by Nick Wareham and reflects Cambridge Public Health’s work to form connections and explore opportunities at the intersection of AI, data science and public health.