Cooperate or Collapse: Emergence of Sustainability Behaviors in a Society of LLM Agents

Emergence of Sustainability Behaviors in a Society of LLM Agents

You might have heard of Google's recent blunders. It’s safe to say that with every instance an AI recommends the ingestion of glue, poison mushrooms, or stones, the more obvious the importance of safety in decision-making for Large Language Models becomes. This research paper aims to address this issue by proposing an evaluation framework (GovSim) that …