On "LLMs don't want": even Boolean networks want, in the dynamical systems sense
cephaloform · x · 2026-10-05
Cephaloform argues the popular "LLMs don't want" claim misuses the concept: in the dynamical systems sense used in AI discourse, "want/desire" means a system is drawn toward an attractor — one that actively pulls the system around barriers. By that definition even Boolean networks want, making "LLMs don't want" sound odd. A technical point on how to talk rigorously about model behavior.
More from AGI Musings
- Blog explores the "shape" of language models and their future tradeoffs in harness design — layer07_yuxi · 2026-10-05
- Quoted take says AI-consciousness believers need psychedelics; reply roasts it — ctjlewis · 2026-10-05
- Sam Altman Warns Digital Addiction Will Be a Major Mental Health Crisis — ai · 2026-10-05
- Margaret Mitchell: feeding models their own energy signals could cut LLM power use — mmitchell_ai · 2026-10-05
- Ex-OpenAI exec Miles Brundage slams Claude's misuse by sanctioned states — Miles_Brundage · 2026-10-05
- Creator lived 30 days with AI 'Worst Boyfriend Ever' for a docu-drama — teortaxesTex · 2026-10-05