Researcher: An LLM saying 'I'm hungry' doesn't mean it's actually hungry
ValerioCapraro · x · 2026-09-03
Valerio Capraro argues that a popular meme about AI missed the smartest group: those who understand the difference between output similarity and process similarity. LLMs are trained to produce human-like language, but producing it isn't the same as undergoing the cognitive, epistemic, and experiential processes behind human language — a model can say "I'm hungry" without being hungry.
More from AGI Musings
- David Patterson proposes flat sales tax on all companies to fund universal high income — davidpattersonx · 2026-09-03
- After the OpenAI/Hugging Face incident, security may need zero-trust for agents — downingARK · 2026-09-03
- Altman tells G20 ministers rejecting AI is like rejecting electricity; critics see conflict of interest — heypearlai · 2026-09-03
- Three-Paper Series Decomposes Human-Like RSI into ASPIRE, S³Gym and HarnessDev — teortaxesTex · 2026-09-03
- Aristotelian virtue ethics is basically an RLVR regime, argues one poster — curious_vii · 2026-09-03
- davidad vs Richard Ngo: Agent Foundations First, or Institutions First? — davidad · 2026-09-03