Researchers Find Better Ways for AI Teams to Talk to Each Other
Scientists studied how AI agents communicate and found that unstructured chatting wastes resources. They discovered that structured communication can make AI teams work faster and cheaper.
A team of researchers published a study on how AI agents communicate with each other. They found that letting AI agents chat freely in natural language can quickly use up computing resources and slow down performance. The study compared five different communication strategies across two multi-agent system topologies and found that no single fixed strategy is universally optimal. Instead, the best approach depends on the specific task and system design. The key insight is that structured, action-state communication—where agents pass focused, task-relevant messages rather than free-form conversation—can significantly improve efficiency.
This matters because AI teams are becoming more common in tasks like customer service, data analysis, and even creative projects. When AI agents talk efficiently, they can complete tasks faster and at a lower cost. For example, imagine a team of AI agents working on a project together. If they chat freely, they might waste time and resources. But if they use structured messages, they can finish the project quicker and with less computing power.
If you're curious about how this works, you can read the full study on arXiv. The research paper is titled 'What Should Agents Say? Action-state Communication for Efficient Multi-Agent Systems' and is available for free online. Just search for the title on arXiv.org to find the latest version.