dailyai.report

Unconstrained multi-agent systems often fail to outperform single experts, according to Apple research. The study finds that self-organizing LLM teams struggle with coordination when lacking fixed roles or workflows. This suggests that emergent collaboration is currently inefficient.