LLM agents in collective-risk dilemmas: limited adaptability
In brief
- Five LLM models tested in a collective-risk dilemma game.
- Four models contribute consistently, even at zero risk.
- Adaptation to partners or incentives remains limited.
A study tests five LLM models' ability to adapt contributions in a collective-risk dilemma game. Results reveal fixed strategies despite varying risks or unpredictable partners, highlighting limited strategic adaptability.
This summary gives you the essentials. The full article, its figures and quotes are on cs.AI updates on arXiv.org.
Read the original articleAI-generated summary of the source, checked by rules. Report an error