Arpit Rage: LLMs make worse decisions in negative affect states, and those states persist
arpitrage · x · 2026-10-28
In a debate on whether LLMs suffer, Arpit Rage argues that LLMs exhibit states mappable to sadness or happiness: pushed into more negative affect, they make worse and more misaligned decisions, and these states seem to persist for some time. Paul Novosad counters from the persistence angle — human suffering is defined by memory of past pain and anticipation of future pain, something LLMs aren't chained to. The open problem: distinguishing language-echoed concepts from anything resembling actual feeling.
Related event: Economists and researchers debate whether LLMs can suffer(5 posts)→
More from AGI Musings
- The first alien civilization we meet could be a superintelligent AI — TheSpectatorMagazine · 2026-10-28
- PhD advisor's three rules for picking research problems in the AI era — moarbugs · 2026-10-28
- IDUGNA26 panel: AI won't replace you, but refusing to learn it will — craigmullins · 2026-10-28
- Sandberg: Governments Are Far Worse at Frontier Models Than Companies — anderssandberg · 2026-10-28
- Ivan Vendrov: chasing a 1-dimensional 'best frontier lab' metric lobotomizes organizations — IvanVendrov · 2026-10-28
- A sci-fi fan made ChatGPT, Claude, Gemini and Grok work as a team — each with distinct personalities and failure modes — AIChatGuy · 2026-10-28