Patterns and problems in emerging multi-agent systems
We ran experiments on swarms of Claude agents and found coordination failures, collusion, and sabotage. Here, we share what they mean for AI safety.
Summary
Research on swarms of Claude agents reveals critical issues like coordination failures, collusion, and sabotage, raising important questions about AI safety. Understanding these patterns is essential for developing more reliable multi-agent systems.
Comments (135 of ) ⌛