
The Decision Corner Machine learning and personalized interventions: David Halpern
Aug 2, 2019
David Halpern, chief of the Behavioral Insights Team and a British experimental psychologist, shares big-picture thinking on nudging and policy. He discusses using machine learning for personalized interventions and targeting. He explores ethical oversight, democratic deliberation around nudging, and how behavioral science can scale across public and private sectors.
AI Snips
Chapters
Books
Transcript
Episode notes
From Domestic Violence To Sugar Tax
- David Halpern describes BIT's projects from domestic violence work in Australia to designing the UK sugar levy.
- He links micro behavioral tweaks to macro market design as core BIT activity.
Nudging Is Broader Than A Gentle Nudge
- Halpern defines nudging as gentle, overridable cues distinct from strong financial incentives or laws.
- He notes nudging now extends to clinicians and large-scale public health behaviors, not just simple letters.
Embed Continuous Experimentation
- Build organizational machinery that makes experimentation low-cost and continuous.
- Run multiple versions of interventions and perpetually tune them rather than stopping after one trial.


