Op-ed: Internal representations don't mean LLMs have emotions
ValerioCapraro · x · 2026-08-27
Responding to an Anthropic paper finding emotional representations in Claude, the author argues that interpreting this as evidence of emotion is fundamentally flawed. It conflates similarity in representation and output with similarity in underlying processes.
The post notes that human emotions involve deep reorganization of attention, decision-making, and processing systems, whereas models currently only simulate outputs. Additionally, since consistent neural signatures for emotions haven't even been found in humans, the attempt to prove LLM emotions via stable signatures is invalid.
More from AGI Musings
- Dijkstra on natural language programming: intrinsically doomed — examachine · 2026-08-27
- Continual Learning is AI's Holy Grail: Fluid intelligence will enable the next capability leap — daniel_mac8 · 2026-08-27
- Game 'Lord of Tokens': Simulating the Full AI Startup Chain — Charlotte1309 · 2026-08-27
- AI Frees Human Time: Community Experiments Rehearse the Future — cccalum · 2026-08-27
- AI is an Intellectual Revolution, Not Just Technological — JoshuaJBouw · 2026-08-27
- Funding the Commons Forum to Address Wealth Distribution in the AI Era — omojumiller · 2026-08-27