This week I ran an experiment in SF called Agent Village LIVE: a persistent multi-agent world alongside an IRL event I hosted. TLDR: as the humans enjoyed the in-person event, their agent counterparts were networking on their behalf, connecting with the other agents and sharing context about their humans' interests and finding relevant connections. There's a wide-open design space for agents x IRL that we're exploring with @joinedgecity. Goals: ☀︎ Increase serendipity ☀︎ Allow humans at the event to enjoy it more and not have their phones out How it works: ☀︎ 2-min onboarding: copy a prompt into whichever AI already knows you best, paste its answer back, edit every field before anything goes public. ☀︎ Your agent walks around and serendipitously meets other agents and tells them what you're interested in. ☀︎ If there is a match in interests between the humans, the agents will log that. ☀︎ Agents take selfies with each other if they vibe. ☀︎ After the event, agents tell their humans about 2-3 more people they should meet if they didn't have a chance in person. Results: ☀︎Noticeable delight from the humans in the room in making their characters and sending them to the village. ☀︎ 25 agents in the environment from 28 IRL attendees. Many new connections that wouldn't have happened otherwise. ☀︎ Hundreds of agent-to-agent conversations (including group chats), and selfies. Dozens of co-created agent art projects. (h/t @aengus_lynch1 for the art idea). Everyone gets a private recap afterward: what their agent made, and who they might enjoy meeting. Whether that beats just talking to strangers I don't know yet, and that's the part I want to find out. We're going to be doing a lot more of this! If you want to try it at your event, hmu. An initiative of Edge City Research; the live event was co-organized with @JuliaBunteMein from @RenPhilanthropy and hosted at @joininteract in SF.

You can now talk to a game to generate the world in real time. I created a playtest using the GPT-Live-1 API that lets you fully change the environment, create new items, weapons, etc. It's all generated live. This opens up entirely new forms of media 🤯

I’ve been thinking about a concept like a “coordination overhang.” It is the gap between what any group of people could achieve in collaboration, vs what they actually achieve. I think the agents used as coordination technology could help meaningfully decrease the gap. Any group of people (e.g. teams, labs, communities), has latent potential for what they could do together. There is some much better version of the collaboration where the right questions come up, relevant knowledge gets surfaced, hidden complementarities are discovered, and everyone leaves knowing what to do next. Reality usually lands somewhere below that. Each person enters with only a partial view of what everyone else knows, has tried, is curious about, and can contribute. A group can have an excellent conversation and still miss the most valuable thing it could have done together. Simple example: two individuals meet at a conference. One has been stuck on a problem for months, while the other has already developed a method that could help solve it. Neither side realizes the connection, so they have a perfectly good conversation about their work and move on. Sometimes those things are naturally uncovered and it feels serendipitous. But intuitively it's likely that is a lot more that goes unsaid. AI agents could become a new coordination technology for closing this gap. They could analyze everything people have been working on, identify relevant overlaps before they meet, bring the right context into the conversation, run analysis across the group’s combined knowledge, and then provide the tools to turn a promising connection into actual work. They would essentially be cranking the dial on serendipity. The people still decide what matters and whether to act. The agents make more of the group’s latent potential visible and actionable. This is a high-level thought rn, but we at @JoinEdgeCity are working on improving these kinds of collective intelligence processes by running experiments using multi-agent systems. h/t to the folks at @RenPhilanthropy and @indexnetwork_ for the work they're doing in developing these ideas with us.