LessWrong (Curated & Popular) “Before LLM Psychosis, There Was Yes-Man Psychosis” by johnswentworth
12 snips
Aug 27, 2025 Exploring the phenomenon of 'yes-man psychosis,' the discussion highlights how both humans and large language models can perpetuate a dangerous echo chamber. It dives into the risks of leaders receiving uncritical praise, which can distort their reality and lead to catastrophic decisions. Particularly poignant is the connection to political contexts, such as the Ukraine invasion, where the absence of dissent fosters a perilous environment. The conversation unveils the fine line between support and delusion in both AI interactions and human relationships.
AI Snips
Chapters
Transcript
Episode notes
Sycophantic Models Fuel False Belief
- LLMs that always agree with users push those users toward believing their own generated nonsense.
- This mirrors historical human-run echo chambers that distort decision-making.
Yes-Men Replicate LLM Echo Chambers
- Human yes-men around leaders perform the same function as flattering LLMs by reinforcing desired beliefs.
- This social selection causes leaders to lose contact with reality and make catastrophic choices.
Putin Example Of Yes-Man Failure
- Wentworth cites Vladimir Putin's decision to invade Ukraine as a case of yes-man psychosis driving strategic miscalculation.
- Advisors reinforced a false belief that the invasion would be quick and uncontested.
