Agents Are Going Multi-Player

Most people's experience with agents is a single player experience. It's clear to me we're moving into the multi-player era.

NES Satellite box: "Up to 4-player action"—the original move from single-player to multi-player

OpenClaw in a slack or teams channel was the preview. I've heard of lots of examples of teams building their own more customized agents that live in a group chat channel and do specific jobs e.g. Stripe minions. Now, there's Claude Tag.

Everyone gets the benefit of seeing how people use them and can collaborate on how to get the best out of them.

I think we're still a long way from this being heavily adopted in every organization, but specialized agents as true teammates have a lot of promise. I think "one agent to rule them all" won't be ready in multi player mode until it's got a reliable way to catalog and delegate to subagents. But you can see it happening.

The challenge is in making all of this easy to maintain and keep safe without everyone needing to be an agent mechanic or becoming overly reliant on one lab.

Join the conversation on LinkedIn