Judea Pearl: I didn't prove smoking causes cancer — I built causal methods
yudapearl · x · 2026-11-09
Judea Pearl pushes back on a claim that he "proved smoking causes cancer," clarifying he instead developed methods for reasoning about causal questions like cancer's causes. He adds that persistent divergences in language between him and statistician Andrew Gelman (@StatModeling) mean no agreement is in sight, continuing their long-running Bayesian-vs-causal-inference debate.
More from AGI Musings
- Zach Weiner: Many People See Abstract Thought Experiments as a Sign of Immorality — sebkrier · 2026-11-09
- zetalyrae teases new post with clickable decision tree of AI futures from ASI to doom scenarios — zetalyrae · 2026-11-09
- Eric Raymond on LessWrong: All Interesting Behavior Is Overdetermined, Not Just Multicausal — esrtweet · 2026-11-09
- Ethan Mollick asks: could AI bring a second 'Special Century' of growth? — emollick · 2026-11-09
- "Being technical" now means having the patience to ask Claude until you get it — brandon_galang · 2026-11-09
- DeepMind CEO Demis Hassabis: one AI-fluent person will outperform an entire startup team — ycombinator · 2026-11-09