Does AI Feel Pain? Dietterich Uses Painkillers as a Litmus Test for Machine Suffering
tdietterich · x · 2026-10-08
In a debate over a paper on whether models subjectively experience pain, Plinz asks what criterion could ever be convincing. Turing Award winner Thomas Dietterich half-jokingly proposes one: painkillers interrupt or dampen pain, so check whether an artificial system reacts the same way. He adds a sharper point: the Conscious Turing Machine (Blum & Blum) models pain but clearly doesn't experience it—so any criterion a CTM can satisfy is insufficient to prove experiential pain in a computing system.
Related event: Turing Award Winner Debates Whether AI Can Feel Pain(3 posts)→
More from AGI Musings
- Peter Yang: AI solved images, music, video — gaming is next — petergyang · 2026-10-08
- 'Corporations are superintelligence' takes get mercilessly mocked — tszzl · 2026-10-08
- Beff Jezos: Physics and math academia became decelerated, only acceleration is the way out — beffjezos · 2026-10-08
- AGI needn't be superhuman: matching an average human on most skills should count — fkasummer · 2026-10-08
- LinkedIn users are retroactively adding 'AI' and scrubbing 'DEI' and 'remote work' from old job listings — sebkrier · 2026-10-08
- LessWrong Thought Experiment: How Should a Model Guess Today's Date With No Date Context? — LessWrong 精选 · 2026-10-08