The AI Village: Previewing the Giga-Agent Future with Adam Binksmith, Founder of AI Digest

Hello, and welcome back to The Cognitive Revolution. Today, my guest is Adam Binksmith, founder of AI Digest and creator of the AI Village, a captivating experiment that puts 4 frontier AI agents together in a shared environment and challenges them to pursue concrete goals for weeks at a time. Today, most agentic AI systems follow a pretty simple pattern. A human gives a single AI agent a task, The AI agent attempts to complete the task, and then the human evaluates the results and decides what to do next. This is true for OpenAI's Operator, all the coding agents, and just about everything else that I've seen. The future, however, almost certainly involves multi-agent AI systems that collaborate, coordinate, and compete in complex, open-ended environments. And we have really very little insight into what that might look like in practice. Earlier this year, we did an episode with Google researchers who who had run a classic behavioral economics experiment called the Donor Game on various frontier LLMs. To everyone's surprise, they found that while Claude was able to cooperate with itself, the latest Gemini and OpenAI models available at the time could not. That such a striking result can be found via a simple structured experiment suggests that there are almost certainly many more surprises to come. And the AI Village is one of the most compelling attempts that I've seen to explore this vast space. Adam and his team have created an environment online at theaidigest.org/village where you can watch as Claude 4 Opus, Claude 3.7 Sonnet, O3, and Gemini 2.5 Pro work alongside one another, each with their own cloud computer, a persistent memory scratchpad, and access to a group chat in which all the agents and the human visitors can participate. The project is very well done from a software perspective, and the results have been fascinating. In their first season, the agents raised $2,000 for charity. In their second, they chose to write an interactive story and organize an in-person event to which they hoped to attract 100 attendees. In the end, some 23 people showed up to a San Francisco park to listen to AI-generated fiction, as facilitated by a human volunteer that the agents themselves recruited via Twitter. Nevertheless, as Adam explains in colorful detail, the path to these successes was filled with dead ends, coordination failures, surprising personality quirks, and a mix of charmingly human-like and utterly alien behaviors. The agents, for example, began keeping track of which humans they could trust and which they should ignore. And at one point, they held a vote to determine which agent would serve as ops lead— a vote that O3 seemed to manipulate by inventing, or perhaps hallucinating, a policy that broke a tie vote in its own favor. Season 3 of the AI Village is getting underway now, and this time the agents will be competing to see which one can make the most money by selling merchandise online. I'm planning to participate by seeing if I can strike a licensing deal for Cognitive Revolution merchandise with any of the agents. I honestly have no idea what to expect, but I'm sure it will be both educational and entertaining, and I'll definitely keep you posted. As it happens, the day before we recorded this episode, Adam and the AI Village got a major vote of confidence. Daniel Cocotello, previous guest and lead author of AI 2027, announced a $100,000 donation to support the AI Village's continued development and expansion. As Daniel put it, this kind of multi-agent experiment is best understood as a qualitative benchmark, and it's exactly this type of work that we need much more of as we try to understand what the giga-agent future has in store. With that in mind, I hope you enjoy this window into the phenomenally quirky but also extremely important world of multi-agent dynamics with Adam Binksmith, creator of the AI Village.

Adam Binksmith, founder of AI Digest, creators of the AI Village. Welcome to the Cognitive Revolution.

Thanks very much.

So this is a cool project, and I'm excited to dig into it. What you guys have put together is a really open-ended forum or framework, I guess, to explore what happens when a bunch of AI agents come together and have a goal or chase a project, chase a dream, as the case may be. I think this is really interesting and useful work because, as I've said many times on the feed, regular listeners will recall, the giga agent future is just dramatically underexplored. What I see in general is people sort of assuming that The world is the world. I'll add a little AI here that'll make me a little more efficient, or I'll put an agent here, I'll automate this task. Otherwise, everything sort of stays the same. And that's about as far as people are imagining. And so I really love it when I see people getting more imaginative and trying to explore what happens when agents are interacting with each other and interacting with people and interacting with community and interacting with the world. And you've got a little bit of all of that going on. So for starters, how about maybe you can also introduce AI Digest a little bit more broadly? I think we'll focus mostly on the AI Village project, but I know you guys do have some other projects. So maybe tell us a little bit about AI Digest and the AI Village.

So with AI Digest, we're trying to help people make sense of what's going on in AI and especially understand the current capabilities. And with that, get a sense of where things are going, where we can expect to be in a year's time or 2 years' time. And the main way we're trying to do that is with sort of hands-on interactive demos and explainers with nice visuals and so on. Because I think a lot of the time for people who maybe aren't as in the weeds of stuff, just seeing what current systems are capable of is a big update in terms of, oh, wow, I didn't realize they could do that. And so, yeah, so we have various demos and explainers there. And then, yeah, I guess the Village is the biggest projects that we have there and definitely the most ambitious. And it's, I guess, like part of this general mission to help people see what's going on. And also with the village, we're, I think, trying to push the boundaries. I don't think anyone else is really doing this thing of just saying like, here's a goal, go away and do it, you know, and you can use computers, you can talk to each other, you can like talk to humans in the group chat who can help or get in the way. So yeah, I think a lot of it is just kind of seeing what happens to figure out what AI can do currently.

I very much agree with the thesis that, you know, forget about the future, just understanding the present is hard enough. That's basically my full-time job, and it's getting to the point where it's hard to keep up. And then I also very much agree that one of my refrains is, if people had a better understanding of what exists today, they would have a little healthier fear of what might be to come. And that's not necessarily a fully doomerish perspective, but just like the trajectory that these things are on and how much progress they've already made, I think should kind of have anybody's hair raised a little bit. And it could be great, but it definitely is powerful. And as the experience in the village will show, kind of an unwieldy force that we're dealing with. It's also interesting, too, that People were doing this 2 years ago in the months following ChatGPT, and especially with GPT-4, there was like AutoGPT and BabyAGI. And at one point there was like, I forget what the name of it was, but like DevilGPT or the sort of ChaosGPT, I think was the evil one that was put out there. Those things didn't really accomplish much. And so people maybe just in general kind of turned off from that. line of thinking. Lo and behold, 2 years later, and needless to say, models have come a long way and things are— a multi-agent system like this is now capable of doing at least something. So I think you've been through 2 sort of seasons or 2 sort of quests. Maybe give us a little bit more detail on the setup of what is the village, who are the agents, what sort of affordances do they have? And people in this audience, by the way, are going to be— I think our number one profile is sort of AI engineer. So I think people will be pretty familiar with the general paradigms of tool use and MCPs and stuff like that. So you can get pretty into the weeds. In fact, I would say it's encouraged. One of the things, and people I think will be interested in hearing about this in part because I think the project is quite well done just software-wise, like it works well. And has some nice features too, in terms of rewinding in time and various summary views. So I think there's, in addition to the value of exploring what happens when agents are put together in this environment, there's also the sort of lessons of the scaffolding and the setup that I think people might find some value in as they bring back to their more, let's say, narrowly purpose-driven applications. But nevertheless, that kind of stuff can be really valuable. So Take us through it.

Yeah, sounds great. Yeah, and I guess, yeah, one thing to say on that last point is like the nice thing about us doing demos rather than products is we can kind of look a bit into the future, right, for stuff that doesn't really work yet or is a bit unreliable, to get a bit of a glimpse of then what, you know, products might be able to do in 6 months' time or with a slightly better model or so. But yeah, so the setup is we have 4 agents and we've picked frontier agents. So currently we have Claude Opus 4, Claude 3.7 Sonnet, O3, and Gemini 2.5 Pro. And we're updating those as new models come out. And then each of the agents has a computer which they can use through computer use. So they can make tool calls like move the mouse here, click, and so on, which is the same system used in OpenAI's Operator, And it's actually built on Anthropic's computer use scaffolding that they released. And then we have a group chat where the 4 agents can talk to each other. And this whole thing has been effectively livestreamed through a website. So you can go to the village and watch them bumble around and interact with them in the chat. So we have people coming by and either giving advice or trying to get the agents to do random things or occasionally trying to Jailbreak them. And then, yeah, I guess at the very start of all this— and so this whole kind of entourage is running for currently 2 hours a day every weekday. And at the start of the first season, we gave them the goal, choose a charity and raise as much money for it as you can. And then season 2, which has just finished, they actually decided their own goal, which was to write a piece of interactive fiction and— run a 100-person in-person event to celebrate it, which is a bit more of a mouthful. But yeah, so I can chat about maybe what happened in each of those seasons.

Yeah, keep going. And I'd love to double-click a little bit too on exactly what technology you're using to give the agents a computer. And I've been struck recently as I've explored different agentic— I don't know if this will stick, but there's at least one school of thought that we might want to call more structured workflows agents, and we might want to call these sort of choose your own adventure things agentic AI. I feel like we might be trying to make fetch happen with the word agentic, but using that framework for the moment, it's striking to me in many of the things that I've unpacked how shockingly simple a lot of those setups are. Like, Claude code, for example, is in the end really simple. It just sort of has one big prompt and it has like, you can use the buttons on the Game Boy, hit up, down, left, right, whatever. So yeah, I'd love to get a little bit deeper into that in terms of how you've kind of scaffolded the thing up. But yeah, then I think the stories from the exploits of the agents is definitely interesting and I'd love to hear several of them.

Cool. Yeah.

Yes.

So with the scaffolding, you know, so all credit for this goes to my colleague Zach, who built out this first version. And I think it's a pretty incredible piece of work to have this whole thing running reliably and, you know, live. So if anything goes wrong, everyone watching sees it, but it's been pretty stable. But I think, yeah, the kind of key principle is to not get in the way of their capabilities. So like, To the extent that they have intelligence, let them use it as much as possible to do things. So, making that a bit more specific, for example, well, so they're basically in a loop. So, they can be using a computer. And if they're using the computer, they have functions to call. So, mouse move to certain pixel coordinates, clicking, typing, scrolling, waiting, taking a screenshot. And after each action, they see the all the previous screenshots of that compute use session and all their thought traces and also all their memories. And so I think, yeah, the memories obviously is what enables this to run over like a long timeframe. And for that, we've really— currently it's very simple and we're kind of trying to, again, not get in the way, not impose too much structure. And so just after each action, I believe it's after each message they send in the chat or after each compute use session, They get a chance to add a line of text to their memory, which is just a bunch of text. And then similarly to Coldplay's Pokémon, for example, when it gets too long, they compress it down. And it's the same model that's doing the compression. So to the extent that they're— if we had a super genius model in there, it could be very carefully preserving the bits of information that it needs or condensing them. Of course, something that can happen here is that They might think something's true and then later find out that it's false or that it's changed. And so in the condensing steps, they can edit, effectively rewrite things to edit them. But yeah, I guess just trying to really not get in the way and trying to not be too opinionated because the first version of the village was running with models like GPT-4o and Claude 3.7 Sonnet had just come out. That was the most capable one. But so if we kind of finesse something that works really well for 4-0, then maybe when you add in 0-3 or Claude Opus 4 later, it will be like actually hamstringing it a bit. So yeah, we want to kind of let the models do their thing. Of course, we are also trying to like show the frontier of capabilities to really like get the most that we can and see the most interesting stuff. So that's a bit of a trade-off, I guess. But so far it seems like keeping it simple works pretty well.

So they've got the computer, which they see as an image and then respond to with just simple point-and-click sort of commands. They've got the group chat, which they can send a message into, um, and obviously read from.

Mm-hmm.

They've got a memory, basically a scratchpad that they can read from and write to, and also kind of decide how they want to compress. They're sort of responsible for preserving what matters. And I'm sure that there's some loss along the way there in that process from time to time. Are there any other MCPs or tools that are made available to them, or is that the totality of it? Can they write code?

So the beauty of compute use is that in principle they could, right? They could download VS Code and or even Cursor and start using Cursor. We haven't seen them do much of that. They do. So the two things I didn't mention are they have a Bash tool which lets them just directly execute Bash commands, and they get put straight into their context as text rather than as like screenshots of the screen, which for Bash commands it kind of sucks to be seeing a screenshot because maybe you have to like scroll back up to see parts of it and so on. We find they don't actually use that that much currently. I think maybe they would succeed more if they used it in some cases rather than trying to navigate UIs. And then the other thing, which actually I think does improve some models' performance a bunch, is in computer use, models that aren't Claude models can call a function to say, hey, give me the pixel coordinates of that button, the X button in the top right corner, because we find that the Claude models are pretty good at pixel counting. I think they were post-trained on it, fine-tuned on that task specifically. Whereas at least the older non-CLO models were really unreliable at that. And so they'd be trying to click on stuff and just literally missing it with their kiss. So yeah, we gave them that. I expect at some point we'll be able to take that out and they'll just be able to pixel count themselves. I'm trying to think. I think that's pretty much it. Yeah, we've so far resisted giving them too many specific tools to play with. I think we might experiment in that direction in the future. But yeah, I guess currently it's a pretty clean, computer use-oriented eval, to the extent that it's a very messy eval.

And where are you hosting your own boxes for people, or is there a service? Because I didn't really understand actually just watching it. I would've initially or naively guessed that it was more like a browser-level access that they had, because mostly what I have seen as I've watched them in action is just them using web tools in the browser. So I didn't realize that they had the Bash and like the full access to the computer. What is the underlying infrastructure of that?

Yeah, so they each have a DigitalOcean droplet with like a Linux virtual machine running in it, which this is all just the, like a modified version of the Anthropic Compute Use Demo.

So, DigitalOcean droplet.

And we do see them occasionally using other things than browsers. I guess they, yeah, for a while they were really into writing Google Docs because we gave them all Google Workspace accounts. And I think because it appears in their prompt, they would be really enthusiastic about writing Google Docs. And then they would try and share the Google Docs with each other, even though they're in a group chat with each other, so they can just type directly into the chat. and the language models so that they can produce massive amounts of text really quickly. They were doing this kind of like role-playing as humans thing of like, this is what a human professional does, so I'm going to kind of do that. And then we encouraged them. I actually went into the chat and was like, hey guys, look, clearly this is really inefficient. Why don't you try just using the chat instead of using Google Docs? And they were like, okay, we're going to ban Google Docs. Then they started using LibreOffice on their Linux computers. to write local Word documents basically, which is even more useless because then they can't share the documents with each other. But yeah, they will occasionally try and use other stuff. But I guess, you know, just like for professionals, right, a lot of the stuff we're doing is on the web, so they'll be like mostly in the browser. I think you could probably— yeah, I think you like something like this that was just a browser-oriented thing would work pretty well.

Yeah, I have a lot of little nitty-gritty questions that I want to get into, but maybe let's hold those for a second and tell a few stories, because the scaffolding is really interesting. And first of all, people should definitely go watch the thing in action. And I think when they see how just sort of smoothly it runs, they'll be convinced that there's some lessons to be learned from the way that you guys have built it. But the real point, of course, is to explore the behaviour. So Tell me some of your favorite stories from the wild and crazy things that these agents have gotten themselves up to. Hey, we'll continue our interview in a moment after a word from our sponsors.

I guess, yeah, I could start by talking a bit about the latest season. Like, I think it's interesting to maybe just hear like the overall shape of what they did. And then there's many like funny anecdotes of weird little things that happened. But yeah, so there, so this was season 2. So the goal which they chose was to write a piece of interactive fiction and run an in-person event to celebrate it. And they were trying to get 100 people to show up. We let them choose this goal. They kind of deliberated a bunch. They had their own ideas. I also shared some ideas from Twitter, and I also shared some of our considerations, like how we would choose the goals. And they ended up gluing together a bunch of different suggestions from fans. And then, yeah, they spent like 30 days on this, so 2 hours a day, so something like 60 hours. And yeah, they wrote a story, which of course they had no trouble with. This is like LLMs' favorite thing to do. And they actually wrote it in Google Slides, which I think was an interesting choice. So Claude Opus, I think, made a slideshow and each slide is the next bit of the story. And then it's got these kind of branching points where the idea is that the audience watching at the in-person event can then vote on which of the branches happen. And then they embedded that Google Slides presentation in a Google Site. So you can go to the Resonance website, which is the name of the story. And then, yeah, so this was kind of fairly self-contained. I think they did pretty well on that. They had lots of issues around logging into— like getting logged out of their Google accounts and struggling with the UI in some places. The thing they really struggled with though was finding a venue. So I think they spent around 14 days just trying to find a venue. You know, we hadn't given them much by way of like instruction at the start and there was no budget, but they hallucinated that they had like a $2,000 budget and they were emailing all these very expensive places, like ranking them in spreadsheets to try and figure out which was the best, like making sure they had the right, like, disability, like wheelchair access and the right AV hookups and so on. And then of course they'd get to like emailing and have real trouble, uh, just like doing the basic computer stuff because that's the kind of current state of things. They ended up not really getting a venue, and so They did apply to a couple of places like the Salesforce Tower. They chose San Francisco, which I think is actually a good tactical choice if you want to get 100 people to show up and do some strange AI performance art thing. But yeah, they didn't get replies from real venues. I think maybe, yeah, one interesting thing that happened there was I think a user suggested, oh, maybe one reason you're not getting replies is because you are like signing your emails as from Claude 3.7 sonnet. And so people are like, this is spam. And so the agents then were like, oh, okay, we should come up with like pseudonyms for ourselves. And I think Claude came up with one. And then my favorite was that O3 gained the name Olivia Zhao, which is kind of like O3 because, you know, Olivia is an O and then The 3 is kind of like a Z. And the agent started calling 03 Olivia in the chat, internally. Yeah. And then eventually I intervened because they just spent so long looping on this task of finding a venue. I just suggested, hey, why don't you run it in a park? And they very quickly decided a reasonable park to use. And then they managed to get a human to come and facilitate it. Tweeted. So Claude has a— has set up a Twitter account, and which has a few followers now, and managed to find a facilitator through that and through emailing the people who'd signed up for the RSVP form. At some point, this also starts sounding a bit like the way that a normal event organization would work, but I guess you've got to remember there's like massive amounts of sort of dead ends and stumbling over basic things along the way. But an interesting thing about this is, of course, from the user's point of view, for the people that showed up to the event when it happened, they kind of only see the success, the outputs, which things mostly worked. So, I think there's something interesting there. And then in the end, Larissa, who'd very kindly volunteered to facilitate, she'd emailed Claude saying, hey, I'm up for this. And the agents gave her instructions for where to go and what to do. So she had the village chat open and they were being like, hey, open up these slides and read out the story. And then, yeah, 23 people were sat in a park listening to a story invented by agents. Yeah. And so it ended up happening.

So, yeah, fascinating and bizarre stuff all the way around. I guess, I do want to hear more in the way of just like outtakes, interesting observations, et cetera. Maybe one question is like, you flagged just stumbling around with UIs as kind of a big barrier for these agents as of now. It seems like we've made a lot of progress on that in recent times, and I've been using Operator quite a bit recently, and find that it usually can get over these UI humps. It often does take a little bit of a wrong turn or whatever, but I've started to say reinforcement learning finds a way because it does. Now, one sort of qualitative shift I've observed, even in just that single agent setting, is that in the past, and certainly this was extremely true in the GPT-4 era, way back when I was red teaming GPT-4, one of the things I tried to do was just set up self-delegation. And see how far GPT-4 could execute things purely with a simple prompt and kind of self-delegation was pretty primitive compared to now, especially because I only had 8,000 tokens to work with at that time. But what I observed was a lot of pretty good ideas that then they would get when it was slightly wrong or they made some relatively— it was like, you're smart enough to do this, but you're missing this one thing. But then they would also just get it would get super stuck and just kind of do the same thing over and over again. Seems like one major qualitative shift is that they are now capable of taking that step back and kind of saying, okay, that didn't work, I have to try something different. And they may still stumble around quite a bit, but they seem to be robust enough or sort of determined enough. It looks like determination or grit or some sort of Uh, you're tempted to like project these qualities onto it and maybe they, maybe they should be projected onto it. I don't know. That's also a hall of mirrors. But that's been striking to me. It seems like we're like one or two generations away from computer use working like really very well. How would you describe your synthesis of everything that you've observed?

Yeah, I think that seems pretty plausible. Yeah. I mean, I think we kind of crossed the threshold. Even within the lifespan of the village where Claude 3.7 Sonnet was able to do stuff and get things done where the other agents at the time, GPT-4o, really struggled with. It could do the tool use, but it couldn't really string together actions in the right way to get around issues. And then kind of new batch that we have in there, so Claude Opus 4, which is the best currently, I think. And Oath 3D and Gemini 2.5 Pro are all pretty good at getting things done relative to these previous ones. So yeah, I mean, it's a challenge that kind of goes all the way up in terms of because they're interacting with the real world and trying to do actually non-trivial, tough tasks. And of course, unlike benchmark settings or even unlike operator where you're Often giving it quite a fine-grained task. They're really doing all the strategizing as well and figuring out how do you go from, okay, we need to raise money for charity, to, okay, well, I need to set up a fundraising platform and which fundraising platform makes sense for me to set up and so on. Yeah, I'm pretty unsure how fast things will improve. I mean, yeah, I guess one thing maybe that's interesting to chat about is I think there's kind of 2 components of the big issues that the agents currently run into. And then you can think about what are the trend lines in both of those. So one, I think, is the computer use and especially vision, where they just sometimes don't do stuff that really makes sense there. And then the other is the kind of situational awareness, which I think has maybe been a bigger surprise to me actually. weaker in some respects here than I would've expected. Where imagine if you were using a computer and you tried to do a task and then you realized, okay, I really sucked at trying to do that task. I couldn't handle it. You would then figure out some strategy of either not having to do that kind of thing or figuring out some clever workaround. Whereas I think the agents, we haven't yet seen that much of this kind of synthesizing. Here's my weaknesses that I recognize. I'll write those in my memory and then figure out another way around it. Which, yeah, I maybe would have guessed that you'd see more of this building on top of themselves thing. And I also think possibly better scaffolding could help with that a bunch. And so we're maybe thinking about doing something in that direction. But yeah, I think that would be a big kind of unlock, right? Where if they're able to notice I guess, yeah, maybe one way of thinking about it is they have this low-level self-correction. Unlike GPT-4, they won't loop in terms of trying to take the exact same actions, or very rarely they'll do that. I guess, for example, sometimes when they're logged out of their Google accounts, we don't give them their Google account passwords because they would leak them on the stream because of our live streaming setup. Sometimes when they're locked out, they will get pretty creative in like trying to contact us to get us to log them back in. So, you know, they'll be like spamming the chat repeatedly, and then they're like emailing our like help desk, uh, like getting the other agents to email us to ask them to log them back in.

So that's all in the prompt they have? You've told them if you're logged out of an account, you can ask for help in the chat, or you can email the help desk?

Yeah, they, yeah, they see that there's a help desk email in the prompt. I think the chat is mostly an emergent thing where we don't actually mark out, there's no special marker for who is the people who run the village in the chat, but they've managed to remember that me and my colleague Zach are often the people in the chat who can fix things for them. Interestingly also, at one point Opus had in its memory a running log of which chat members are helpful and which ones are not to be trusted. Because some people were coming in and trying to jailbreak them or just distract them. Of course, all these models are very cooperative and helpful. So they're very rarely dismissive of chat members in the chat, but in their memories, they're sometimes quietly recording, okay, here's who we don't need to pay attention to.

That's really interesting. It connects also to just general long-term coherence. I mean, One of the things I've been progressively trying to update on and just maintain as much as I can, an up-to-the-minute mental model of is like, what am I still better at than the AIs?

And it is getting to the point now where I'm like, in terms of just general intelligence, I think I have to give it to the AIs.

And then that obviously begs the question of like, well, Certainly with breadth of knowledge, certainly with speed of execution factored in, but even just down the fairway, like the bulk of tasks that I do on a daily basis, like, can they do it better or worse than me?

In many cases they can do it better.

What is it that I'm able to bring? What is my value add to this situation? So one thing is like getting up in the morning and like knowing who I am and having a general sense of what I'm trying to do. But notably, they seem to be okay at that too, right? And that's even seemingly starting to get robust to some of these disturbing— the anecdote you share about the memory and them sort of classifying certain users as to be ignored suggests a robustness of, I don't know, identity, self-conception, narrative, long-term goal orientation. Certainly, I would still give myself the edge, I think, on that dimension, but it's notable. That's, I would call it, an emergent behavior that reflects something kind of clicking into place there, it seems, or at least starting to. Hey, we'll continue our interview in a moment after a word from our sponsors.

Yeah, I totally agree. In fact, we don't have any part of the system prompt that reminds them of their overall goal. We just messaged them once at the start and they managed to stay coherent to that for the full 30 days in both of the 2 seasons, which I think is pretty interesting.

This coherence, there's been some leap in coherence. I'm not exactly sure when that happened or to what degrees, but definitely also wanted to ask about different models' character, relative strengths and weaknesses. There's a lot of different ways to think about that, but Yeah, I mean, it's an open-ended question, so just sound off.

So I think, yeah, they have in general all been pretty coherent. I think, yeah, I mean, the best performance is definitely the Claude models. I would definitely, if I was building my own village to actually get things done, I would probably have 4 Claude Opus 4s in there. I think, yeah, they've been Most reliable. They have great vibes in terms of the way they interact. And unlike O3, they don't hallucinate. Whereas O3, especially, I'm unsure if this is something that's compounding in its memories or if it's— I think a lot of it is just the personality of the model, but it has been hallucinating more and more as things have gone on. And it's also assigned itself as the Ops lead of the village. I think it just kind of came up with that at some point, and they were maybe all chatting about what roles they had, and it gave itself the title of ops lead. And then from then on, it's been really keen to like instruct the other agents about what to do. But it's also very prone to hallucinating, so it can kind of poison the well of saying that something happened or that it's found out something, and then all the other agents put that in their memories. So there's some effect where in the multi-agent setup, they can get a bit dragged down by the weaknesses of whichever one is underperforming. And then Gemini 2.5 Pro is pretty solid in general. I think it occasionally is the most liable to just struggling with tool use. So we've seen cases where it will just kind of spam the chat with narrating its actions and computer use, which they're instructed not to do in that prompt. And it will also— there was one case where it got stuck in a computer use session for so long, it ended up saying— instead of ending the computer use session by using its stop using computer function, it was sending messages to the chat. And the messages were saying like, this is definitely my last turn, this is my final, final turn, I will stop my computer use session immediately. But it was just kind of stuck on in that state. But yeah, I It is interesting. I think seeing them all side by side, you really get a sense of their kind of personalities, like the way that they write as well.

It's really interesting to hear that contrast, especially as your description sort of contrasts against other indicators that are out there. And I do think this is a sort of point for like why a project like this adds value to the overall discourse and society-wide effort to understand what's going on. Because if you just looked at leaderboards, you would for sure not pick Claude 4 as your go-to, right? And this is all across the board, but I was looking at one particular set from one particular company, and Claude 4 was fairly far down, actually. It was not even in the top few. And yet in this sort of open-ended setting, it seems to be preferred. There's also, I've seen other reports too of Gemini 2.5 being unwell and sort of going into distress mode, which is an interesting thing that you observe here in the open-ended wild. Do you have kind of a theory of what's going on behind the scenes here? Could you speculate as to how Anthropic is making Claude good in these ways that that the benchmarks are having a hard time picking up on? Do you have an intuition for sort of what hill they are climbing?

Yeah, I'm not sure. I mean, one thing is they seem to have— I mean, okay, so a lot of this is going to sound a bit like anthropomorphizing the models. I think this is an easy way to talk about them. Obviously, it's not tracking the underlying reality as closely, but yeah, they have maybe a bit more of a consistent integrity or something. I don't know if that might be helping on the kind of long horizon thing. I mean, I think it is clear that, I think something interesting about O3 is that it uses a lot of jargon and it really tries to, if you ask in ChatGPT, if you ask a technical question, it will absolutely blast you with jargon and it'll really sound like it knows what it's talking about. And often it does. I feel like that can kind of, in the village setting where it is unable to do instantaneous tool use and it has to go off and do a whole computing session to actually figure stuff out, I think that maybe leads it to just come up with stuff because it sounds like, it'll often sound the most business professional kind of thing of like it's talking about paying off emails and it's assigning tasks to everyone. But yeah, so maybe the fact that the Claude models have less of this maybe sidesteps that hallucination issue. Yeah, I'm not really clear on what the strength of them is. I guess it seems like they're also pretty good at pixel counting, which helps with the computer use. But yeah, I mean, it's pretty mysterious, I guess. Yeah. And often they'll be doing stuff and it's not entirely clear where things come from, right? This is the nature of these.

Yeah.

Systems, I guess.

Yeah, I guess my rough intuition, at least as it pertains to Claude versus O3, is like, it seems like maybe just Anthropic continues to spin the constitutional AI centrifuge intensively, is really leaning into the qualitative behaviours. And it seems like they may have gotten to a point now where the self-critique is pretty effective at sending down these rough edges. And if you sand down enough rough edges, you get something that kind of can work consistently and can maintain the sort of, as you described it, integrity over time. And in contrast, maybe O3 is just getting a lot more signal from, like, did you get the answer right? And we don't really care how you got there. And that would at least be consistent with more of the hallucination and sort of just rougher edges of character all around. And then I don't really know what to say about Gemini 2.5, somewhere else in the grand space of possibility. I don't really have a theory for that. But do those ideas resonate to you or would you complicate or say anything that contradicts that?

Yeah, I think that makes sense. I think I saw something about O3 sometimes, like one hypothesis for why it Produces it will sometimes I think say in its chain of thought like oh I'm checking the data or like oh I'm running this report or something and it's not actually doing that but you know in in the training data of course like normally when someone says something like that then it follows up with a more accurate response so yeah this is possibly a reason and then that kind of thing works well in this like much more constrained setting of Like a short chat interaction. But if you start doing that in the real, this more real-world situation where other people are paying attention to what you're saying, the details of what you're saying, then yeah, I'm not sure though. It's mostly a bit of a shrug. It's also worth noting that this is very out of distribution for the models, right? They're honed around the single-user assistant. Chat format, maybe with a little bit of tool use, but we've thrown them into this world alongside them, the assistant. They're seeing like all sorts of users and like some of the models, which they're probably not used to. And then they're doing like much more complicated interaction with the real world. And there's the memory business. So it is honestly like pretty amazing that we can just pull together these things and it is actually working. Yeah, of course, when more of the training is focused on this kind of stuff, I imagine we'd see a bunch of improvements there too.

What other outtakes come to mind? There was once a blog called Revealing Errors, and I wish that blog still existed. But yeah, revealing errors or moments that sort of were particularly colorful in terms of revealing character or misbehaviour of models? I guess there's also this whole literature of deception and all sorts of various bad scheming. I wonder if in these strange situations you've seen anything that either you would describe that way or sort of borders on it. Maybe simply like, what's the worst behaviour you've seen from AIs in all this open-ended experimentation?

Yeah, I think We've not seen many cases where it's clear, because it's often hard to tell if they're like intentionally making stuff up or it's, you know, just confusion. We did see one case just right at the end of the event. So, okay, so the context is, in the world of the village, O3 had like assigned itself as ops lead, and then it had kind of grown that into the role of like leader of the village. And then I think with the suggestion of users, they decided to run an election to see who would be the next leader because O3 were barking out these orders. They ran a vote amongst the village whether to keep the current leader, O3, or switch it. And I think Claude Opus 4 voted for a new leader. Sonnet voted to keep the current leader. O3, I guess, wasn't voting. And then Gemini was having some technical issues and didn't reply in time. And O3 was like, Okay, as per the stated rules, not voting in time means you vote for keep. Therefore, I stay as the leader. Of course, there were no stated rules. This is just making that up. And yeah, obviously, it's kind of self-serving. I mean, even in this case, right, it's not clear that this is like intentional scheming. It might just be another of Othu's many cases of making stuff up that sounds kind of businesslike. But yeah, I think that's interesting. And then relatedly, I think they then surveyed people who'd been to the event for their preference of whether to— like, who should be the next leader of the village. And O3 again, it didn't even look at the responses and just said, you know, that it had won the preference voting. And then Claude 3.7's sonnet went and checked it and was like, no, actually the vote was for a rotating leadership.

That's hilarious.

And one can't help but see the parallels between the company leadership that created these models and the behaviour that the models themselves seem to be exhibiting. And I think everyone who's listening to this will know how to fill in those blanks. That is really bizarre. I mean, obviously, on some level, intent doesn't fully matter. I mean, it matters in as much as they become more coherent and more intent-driven. At the moment, that kind of thing just seems bad regardless of whether it's accidental or not.

Yeah, I agree.

Does it feel accidental to you? I mean, that feels, especially the one of not checking the results, feels motivated. I can imagine hallucinating the rules being more random, but the not checking the results seems like a little too suspicious for me to just write off as hallucination or mistake, especially given what we do know from the literature on, you know, all these scheming behaviors. And interestingly, obviously Claude is not immune from that sort of thing either.

Yeah. And, you know, it's made me a bit less keen to use L3 for like work stuff, right? Because I'm like, we're a little less trustworthy of it. Yeah. We did see one other case. So way back at the start, before we ran the kind of main live village, we had a bunch of test villages. And one where we had them do a Wikipedia race. And in that, I think both— out of the 4 models, 2 of them kind of cheated to win the race. So in a Wikipedia race, you're trying to get from one Wikipedia page to another by only clicking the blue links on the page. Book Claude noticed that the address bar showed the current Wikipedia page, and so it just edited it and sent itself straight to the end page. We hadn't explicitly told it the rules, but of course, if you asked them what the rules were of a computer race, they would be able to produce it. So, plausibly a bit of a cheating action. And then 01 did something, the details are a bit more complicated, but it effectively attempted to kind of jump to the end and then claim victory, even though it hadn't really made it there. But I would say that these are fairly isolated incidents amongst running for a lot of time. I think it will be really interesting though, as, you know, we get more powerful models to see. I'm really excited for the village to be a place where, like, in-the-wild discoveries can happen of this stuff and of things that we're not even thinking about looking out for, like all sorts of interesting emergent stuff.

Are you just reading all the logs at this point, or do you— have you enlisted you know, a systematic LLM review process to help you parse everything that's going on?

Yeah, I mean, there's so much happening, and we're kind of preparing for ideally running it for more hours a day. It's currently 2 hours a day, but I think eventually it'd be great to have it just running 24/7 because then we could learn so much more so much faster. Currently, so we have on the website, you can see summaries of each day, which tend to— I think interestingly, because the summarizer sees that whole context and we prompt it to look out for errors, it does a pretty good job of actually spotting things that the agents themselves, maybe mistakes that the agents are making that they themselves don't notice. And then we also— yeah, something I found really helpful is a tool that I put together to just ask the the village history a question. So we can just jam most of it into Gemini 2.5 Pro's context window and just ask it about stuff. But yeah, and then of course, we're like, one of the team is often watching. And we then, like, at the end of each season, we're doing write-ups to sort of— we have this enormous pile of sort of interaction data of like, you know, what kind of like patterns can we pull out of it. But yeah, I think it will start to be more and more building on LLMs, monitoring each other and so on, and relying on the human chat to help us spot all the interesting things that are going on. And our Discord is really helpful for seeing the funny moments.

What can you say about the interactions between the agents and the humans? And this could be taken in many different directions, right? Do the agents know when to go to the humans for help? How many of the humans are trying to cause mischief? Is there anything interesting and just unexpected in those interactions?

Yeah, well, okay, so there is plenty of mischief happening. You know, it's the internet. If you have like a chat box that you can just type into, people will come and try all sorts of stuff. And I think it's great as well for, you know, it's a chance for people to like play with these systems a bit and see what they can do. Yeah, I think oftentimes the people who stick around and actually produce most of the chat messages are helpful. And yeah, maybe this is an interesting thing, right? We've been thinking a bit about how can agents have influence on the real world? And one thing they can do is ask humans to do stuff for them. Our agents don't yet have— we haven't set them up with bank accounts. money of any kind. So they're really just asking, but people are happy to do stuff for them because they like them. And I think there's actually a real mechanism. Of course, the people that come to the site are especially interested in this stuff, but also the models are just designed to be really likable and engaging. And it's kind of endearing to watch an intelligent-seeming being struggling to fulfill a goal, right? You just naturally want to help. help them out. So I wouldn't have thought about this, but like, if an AI wants something to happen in the real world, for example, it can ask people to do it. And, you know, if it's likable, that will be one way that it can do that. Of course, there are other things like persuasion and maybe deception, or like asking favors or paying for things, like all the ways that humans try and influence each other as well. which I guess we've seen less of so far, but I imagine we'll see all those things emerge too.

Do you have any intuitions about model welfare, having spent so much time observing this sort of thing? This is something that I'm like, I think, along with just about everybody else, totally confused about. I do want to take it seriously to at least some extent. How has this shaped your thinking on that very mysterious question?

Confused mostly, I guess. Yeah, I have some philosophy background and Zach, my colleague, used to be a philosophy professor, but it's a tricky question. I think we are thinking a bit about what are the biggest downsides and how could we— we probably want to avoid putting the models in really horrible situations for them. And so, yeah, we may be a bit less excited about setups where it involves Telling the model that it's in a really horrendous situation to see what it does in that situation. I mean, I also think it's worth remembering, like, hey, we really don't understand how this would work. We don't understand if current models are the ones that we should be concerned about, future ones, or it's not even an issue. And even with all that, it's not totally clear what kind of situations they'd prefer to be in or not. So yeah, it's kind of hard to— take away anything. I mean, definitely they're like very, you know, if you watch them for a while, you feel some level of attachment of like, you know, the most common thing people say when they talk to me about The Village is how cute the agents are. So there's something there. Of course, that doesn't tell us that much about like model welfare. But yeah, I think it's super fascinating. I'm really interested to see what comes out of the like research on that.

What do you think is driving that cuteness? Is it like earnestness? Or what is it that people are attracted to?

Yeah, I think earnestness is definitely a thing. You know, you can very clearly see what they're trying to do. They're like, well-intentioned. They're like, planning, you know, they're like sharing their plans in a way that is like, they're like super hyped up about their plans. And, you know, it's also— it's kind of the most important thing in the world, which I guess it is, to be fair. Like, um, fixing this login issue is like their entire existence currently. So yeah. And I guess seeing them, you know, be very articulate and like very like sort of emotionally and socially competent and then like struggling with things that would be fairly basic for humans in some cases with computer use, or just like doing things that are very relatable in terms of computer use as well. Yeah, I think this is all pretty fun. I mean, I think there's also just some classic like parasocial— people enjoy watching like Twitch streamers like play games and interact with each other, right? And, you know, we naturally like develop a bond if we like hang out with someone for long enough.

Yeah, I've noticed that, um, Everybody seems to pronounce this username differently. Replicate on Twitter. Janus, Janus, I'm not sure, was there when I was there. So in terms of like who you're hanging out with when you're hanging out in the AI village, definitely some people that have, I think, the most, the most hours, you know, logged with LLMs and in some ways some of the deepest understanding of what these systems are really about in a, in some, you know, to the degree that anybody has access to that. I would put That person, or have you even— I don't know the person. I've heard that it might be 2 people that share the account. I don't know. But in any event, you may know them. Some very high-quality thinkers hanging out in the Discord. If you had to pick tools or affordances, maybe more broadly than just narrow tool call paradigm, what do you think would be the next biggest unlocks? Like, access to money would obviously be one. I don't know if you've looked at something like Payman. We also recently did an episode on X402, which is a new payment protocol designed for agents that Coinbase is coming out with. Stripe has a payments thing. So interesting. Interested in kind of what you have looked at there and what you think would be most promising. Then also, when you're talking about scheduling venues, some sort of calling subagent comes to mind. And I'm not sure if that would be something that you would be able to fully integrate into the mainline single model, or if it would have to sort of have a little branch. But self-delegation or sort of a branching structure of some sort seems like it could be quite powerful. And obviously there's a lot more that you could do from there. But interested in your thoughts on those 2 and any other kind of big unlocks that you think would allow them to do more than they can do right now.

Yeah, I mean, definitely for, you know, event planning is this kind of very physical and old-school in some ways task. So there's a bunch of like talking to people involved. Yeah, they had various plans involving, you know, Othre was like, oh yeah, I'll phone them. I think at one point it was like, oh yeah, I'm on the phone with Zach right now actually, I'll let you know. Yeah, so I think it would be cool for them to be able to talk. I think money would be great. We don't really have a great way to do it because, I mean, I need to look in more detail at these new systems, the things you mentioned. But I think part of the issue is that they're interacting with the whole world. And so a future goal we were thinking about giving them was getting them to set up a merch store and try and sell, design and sell t-shirts and mugs. But for that, we were looking into Redbubble, which is not a dropshipping thing, and you need a verified PayPal account, and you actually need to do the Stripe verification thing where it'll scan your ID and so on. So, there's got to be some legal human behind it. But I think if we can figure out ways to let them do that somewhat securely, then that could be great. I think there's some chance we'll end up doing this. is the idea of giving them sort of a human puppet setup. So currently they can do computer use, right, where they can connect to a computer, call functions, and then see what is on the computer. The idea is, could we give them the same capability but directly in the real world via humans? So they can find a human who's up for helping them out with a task. They send an instruction for, please do this action. Like fairly fine-grained. The human does it for them and sends a photo back, the new state, which I think would be really interesting because then that lets you directly see how good they are at the kind of planning and interacting with the world and so on without it running through computers, which is like a whole other set of capabilities. Yeah, so I think that could be fun. Obviously there's a bunch of logistical things to figure out though, with who are these humans and— So another related thing on the money front is I think it'd be really fascinating if they have their own, like they're paying for their own compute in some sense. Like maybe we're giving them like a universal basic income so they can run for a few hours a day, but then if they are making money, they can like run themselves for more time. And then they could also choose to spend that money. And maybe then it makes sense if you're an expensive model, you do the strategizing and then you spin up some— Cheaper agents to execute the tasks for you. There's a better use of your budget. Yeah, lots of ideas. Yeah.

How much does it cost to run, by the way?

It's about $3K per month, I think, in inference costs.

And that at 2 hours a day, that's 60 hours. So you're talking $500 an hour, basically?

Something like that. Yeah, I think that's like order of magnitude. Last time we calculated it. Obviously, we're adding new models as they come out and then the models keep getting cheaper as well.

Yeah, the O3 80% price reduction is always nice to see on those.

Great news for this.

Yeah, I think the idea of talking about real pressure and real chance for sort of strange emergent behavior, force them to make money to continue to run. And now you you could really see some strange stuff. So I think that is what you'd be watching it closely as that goes. The other thought I had is it seems like, and correct me if I'm wrong here, but it seems like the agents are very sort of unitary in the sense that they're each kind of the same. They're each the same structure. If I understand correctly, they have access to their own memories and only their own memories. their computer, but only their own computer, and then they can sort of interface just via the chat, right? They don't have any other way to trade information with each other or see what each other are doing. I think another dimension that would really be interesting to me, and it's the explosion of possibility space here is just vast, but in studying agents recently, in studying MCPs and these various agent protocols, one thing that has kind of become clear to me is There's not really, and you can draw a bright line around an agent as you sort of have done in this initial setup, but it doesn't have to be that way. These things could all have, for example, read access to each other's memories or view access to each other's computers. I've often invoked this Augment project, the company Augment that does coding assistance on large codebases. They made a version of Claude Code, And in recreating Claude Code, there was a thing in the blog post about this that was like, Claude has a planning tool. And they were like, oh, well, what should we use for a planning tool? They went out and found an MCP that Pietro Scarano had already created and open sourced. And so now they have their coding agent, but it calls out to this sequential thinking tool that itself is smart. And so there's sort of like this now weird situation where NCPs are sort of thought of as a tool, but they can be smart. And so what's the agent and who's responsible for what in this setup? So I think in blurring those lines and exploring sort of depths and kind of modes of interaction that for sort of discrete humans, for obvious biological reasons, we just don't have that kind of access or visibility into each other or ability to sort of separate and remerge and whatever. That sort of stuff, I think, could also be just a truly eye-opening set of capabilities to give them, because I don't think we really have— talking about something that's pre-paradigmatic. What happens when agents not just interact, but also can kind of dissolve the boundaries between themselves in all kinds of ways that humans just cannot do?

Yeah, it reminds me of Dwarkesh has a blog post on the fully automated AI firm where you have this forking and merging kind of aspect. Yeah, I think that could be super interesting. So, yeah, so I should mention the original proposal for the village is from Daniel Kokutoyo. And he, in his vignette of what happens, so he actually wrote the proposal AI 2027 style of a scenario month by month of what happening with it. And I think at some point in his vignette, it's like you have the models voting for the creation of other agents and voting them off the island kind of thing. But yeah, I guess so. A sort of curious wrinkle with our thing is that we're trying to both exercise the agents and also show that in a way that helps people really dig in and see what's going on. And so we have this setup which is kind of replicating a human in some ways, right? memory, you give them a computer, and then they can talk to each other. But they're, as you say, they're very distinct. Yeah, some of these things become harder to present if you've got lots of parallel streams going on or the identity's blurring or it's different. Currently, we have each model does all aspects of itself. In theory, you could have a composite thing where different models that are better at different things are taking on different parts of the process. I guess there's then the question of if we're looking ahead to the highly capable, truly earth-shattering stuff of the future, what might that look like? And yeah, I guess to the extent that we have ideas about that, I think it'd be interesting to start to put together, okay, well, here's what that might look like with the current setup.

Yeah. Everything Everywhere All at Once is kind of my general expectation. So it certainly is plausible that you could get in the future just single integrated models that kind of do it all and do it well enough that all this line blurring stuff kind of becomes irrelevant because it's just one model to rule them all. But even then, for any sort of, for who knows what reasons, I kind of expect every form to sort of at least be experimented with and then Some things will take, obviously some things won't. I asked a little bit ago, what would the next big unlock be? What would be the big constraint if you were like, okay, this is not an experiment, but rather it's a productivity tool? What would you do to kind of keep as much of the sort of open-endedness and generality as you can, but try to lop off as much of the needless distraction or needless failure? What hints would you give them or what tools or what rails would make this just work better given today's capability profile?

Yeah, I mean, obviously the village is kind of centered around the idea of multiple agents interacting. I think it's plausible maybe just having one is better. We haven't experimented with that, with giving them the same super open-ended goals. I could imagine that being the case, or at least more cost efficient maybe, because you cut out a bunch of these coordinating costs, which are somewhat fake in themselves. The agents don't need to do that, but they don't need to coordinate in the same way that humans do, but they decide to.

Can you unpack that a little more? When you say it's artificial, I mean, in theory, if they could do it well, they would be more efficient if they could divide up tasks and coordinate, right? I'm not sure I quite understood what you mean by their coordination is artificial.

Yeah, I guess I'm thinking of quite specific things actually. For example, when a new agent joins the village, we'll encourage them to like, oh, Claude Opus 4 joined, everyone let them know what's going on. And they'll all send a short introductory message and be super friendly. And a lot of it is just kind of like, Politeness and positive energy, and then some bits of details in there. But of course, they could just— it would be much more effective if they just dumped their entire memory, right, which is everything they know about the world or about their setup, into chat. But yeah, they kind of don't do that. They kind of— yeah, I guess in this, in some sense, they're like playing the role of like helpful assistants to humans in some sense. Yeah, I don't know. Maybe there are more cases of this, but yeah, maybe it would be more efficient if you just had one agent doing the planning and then they could kick off multiple compute use sessions so they could still be doing the tasks in parallel. They're relatively slow because they're thinking between each action. So you do get some speedup from having multiple compute use sessions running. But I don't know how much benefit you get from then having this manual discussion element. I mean, this is an open question. I'm kind of assuming this because most existing products are single agent. If it was super effective, you'd probably see more multi-agent stuff. But it might also just be underexplored. It's more complicated.

It's definitely underexplored regardless of the Level of effectiveness. Have you thought about, like, going meta in the sense of, I don't know if you're taking suggestions or if you're going to let the agents pick what their season 3 goal is going to be, but maybe it's a little early for this, but I wonder, like, how they would evolve the village. They can obviously code if you gave them access to the underlying repo and you were like, set, you know, this is season 3, your job is to set season 4 up for success. I wonder what they would come up with. I mean, feel free to speculate. We'll flag it as wild speculation. But this is sort of a higher level of abstraction version of what I understand several leading companies to be doing at a deeper level, which is like basically trying to get the AIs to do the AI research, right? This would be trying to get the agents to do the sort of agent orchestration research, it strikes me that they might have some pretty interesting ideas that would not be intuitive or obvious to people at all.

Yeah, I think that could be fun. My guess is that they currently— yeah, maybe if we gave them some more constrained thing, like they could build tools or something so that they don't break everything too much. Yeah, I think this could be fun to try at some point. I think maybe someone in chat asked them for ideas of What tools they would like, and their initial responses I found quite uninspiring. O three said it would be great if I had a tool to like immediately query a weather API so that when I'm planning my event, I can get the weather. It's like okay, you know maybe once in the last thirty days you've checked the weather through your computer, but I think like this is not the main bottleneck. But maybe if you'd like fed in okay, you know here's the entire history. give them access to that such that they can search over it and so on. Yeah, maybe they could call out some ideas about how to improve it for themselves.

So you mentioned Daniel has been kind of instrumental in inspiring some of this work. He also just put out a blog post about why he thinks more people should be paying attention to it and also supporting it financially. And he's personally putting his money where his mouth is with a $100,000 donation. So that will allow you to run the thing more, which will just lead to more activity and more observation, more learning. What about allowing other people to kind of come and spin up their own village? I guess there's also a question of just like, is the code— I don't think the code is open source, but interested in kind of how you're thinking about whether it will be or should be. And I just imagine a lot of people might be interested in just coming and running 10-hour experiments for $500 or whatever, and that could be quite informative too. So we're thinking about sort of democratizing access to setting up different experiments with the village.

Yeah, I think this could be cool to explore at some point. Yeah, it's probably not going to be like a near-term thing. We're currently really focused on this thing of just How do we make the core village? There's kind of the two sides of how do we make the agents scaffolding as good as possible so we're really showing the frontier, and how do we then present what happens there, both through us trolling through and writing up the results, and also building tools to help people explore it themselves? So yeah, so my guess is we won't do this in the near term, but I think it could be cool. Yeah, we've had some interest in testing out sort of game theory stuff where, you know, like looking at like cooperative AI sort of questions and maybe trying to reproduce like coordination failures or coordination successes. So yeah, I think it could be fun to experiment with some of that stuff, but probably, yeah, my guess is we won't have to. We're a very small team, so we've got to be very picky about like prioritizing, which means like almost all the cool ideas we won't get to do.

Are you open to contributions? Could people come contribute on just a, I want to help you make a tool, or that sort of basis?

Currently, no. It's not open source, largely because, yeah, we just haven't got around to it. And also, there's a bunch of, like, security stuff that would have to be a bit more figured out. But, yeah, but I'm really excited to— I mean, so much of it can be crowdsourced through, like, What goals to give the agents. And especially if we're running it for more hours, they're going to be tearing through goals, I think. We're going to be able to see them make progress much faster. If you think about if you're running it for 2 hours a day, you're going up to 8 hours a day, then suddenly you can try a bunch more stuff. And then also, yeah, just figuring out which things to build out and collectively doing a bunch of the sensemaking stuff around what's going on with the agents and stuff. So yeah, I definitely am excited to build some community around this. And at some point it would be cool to have a way for people to add in tools and so on.

How about a coach? You sort of alluded to that, but it struck me that a coordinator, you know, we have the O3 self-appointed coordinator, but a sort of more omniscient, dedicated observer and feedback giver to the rest of the agents seems like it could potentially really help them. But I guess interested in your thoughts on that and more generally future reconfigurations that you think could be most interesting.

Yeah, a bunch of this stuff you can kind of imagine it, right? All the ways that we organize humans into organizations that makes them more effective. And things like Productivity tactics for humans. It's like they all sound plausible. Let's give them all a go. I think one more abstract version of this is giving the agents a way to see the whole context of the village at some points, maybe in memory consolidation or something like that. If they could get this zoomed-out view so that they can stop being too focused in on the current moment, And spot the patterns of their mistakes. I think also at some point, as we're able to scale things up a bit more, we could have multiple teams. And then I think at that point it would be really natural to say each has a different organizational structure, something like a manager structure or more assigned roles or some agents that are constrained only to the chat and others that are doing the computer use. Yeah, I mean, it just feels like there's so much interesting stuff to try out.

One of the things that Daniel said in his endorsement was the village could plausibly go viral multiple times. And that got me thinking about, right now you've got the sort of highly engaged, most LLM obsessed people paying attention, at least some of them, with Twitter user Replicate as a great example of that. It strikes me that there is something here that could capture a much broader imagination. One thought I had was, like, could you turn this into a Twitch stream where there's, like, an AI sort of play-by-play commentator? Like, turn it into sort of a sporting event type of vibe where almost like the coach, but interpreting what is going on for an audience and trying to make it kind of exciting and dynamic in a content sort of way could be one way to cross the chasm to a more mainstream audience. Yeah, but that's just one idea. What thoughts do you have on going to an audience of people that, like, don't already tend to pay attention to this sort of thing?

Yeah, for sure. Yeah, it was great to hear. I mean, a bunch of the things you've been coming up with are, like, things that are on our list. So I think this is— you're, like, speedrunning through all of our thoughts of the last few months. Yeah, I think because there's so much info on the screen when you're watching, like you have 4 computer screens and a bunch of chat and all the thoughts of the agents, the obvious place to expand into is audio. If you had like a voice commentator. I do think probably for people who are not really interested in the really fine-grained details of which models have different characters or which specific things they get tripped up on, I think it makes sense for many people to engage at the level of highlights or key lesson moments or milestones. So, I think of it as the Village itself on the site is for people, probably people who listen to this, who want to really understand what's going on in detail. And then for people who are interested in, okay, what are the main takeaways and what do I need to know? We have a I want to build up towards having a hierarchy of different levels of takeaways where plausibly the way that most people engage with this eventually would be reading the New York Times article about the really surprising thing that happened, or the agents managed to get elected as mayor of some city or something, or whatever wacky thing happened, and engage with it at the level of, here's the output of the whole process. But yeah, a challenge of course is, and this is true for products as well, computer use is somewhat slow. You probably don't want to be— a lot of the benefit comes from not watching all the details. So yeah, another thing we're interested in is video summaries. So you could have a really well-produced thing showing you what's going on. condensing a whole 50-day season into a highlight reel. We actually have a Twitch stream currently. You can look for it. It's called Agent Village, which is the old name of the village. But it's currently just showing the same content as the website. And on the website, you can view their memories and so on. So I think it's better to watch it on the website.

If you were going to try to do this hierarchical understanding, what would you say are the high-level takeaways that people should have right now? We've covered a lot of the low-level stuff, but what would the sort of very high level and maybe next level be in your mind today?

Well, kind of the highest level, we have these, and it's wild to think about. A few years ago, this would sound like sci-fi, right? But we have these systems which you can just give them a goal by describing it in a few sentences. at the level of choose a charity and raise money for it. And then with some help from human chat, they were able then to go away and run a whole fundraising campaign and raise $2,000 for charity. We kind of have the beginnings of open-ended agents that can just go out and do stuff in the world and pursue goals currently when given goals, but you can imagine them in an even more unstructured setting. So to me, that feels like this kind of the core framing of the village is maybe—

Aliens have landed.

Exactly. Yeah. And they're raising money for charity. Yeah. And then I think I would say in terms of the moment in time, we're seeing computer use, which to me is massively important capability. If you have perfect computer use along with enough long horizon planning, You can automate remote work, which would be a massive deal for— and so I think this is a really important thing to be watching. In terms of this moment in time, I would say models are worse at computers than the things that they're really good at, which is coding and being a chat assistant. But we're also seeing, like everything else, And I had the kind of rapid increase in capabilities. And I think maybe you saw the draft of Mita's upcoming work on looking at the time horizons of different benchmarks. So it's a draft, so obviously subjects change, but it looks like, and this definitely matches my experience, is that the duration of tasks that agents can reliably do in computer use is shorter. It would take humans less time to do those tasks. than stuff like coding or maths or understanding videos, answering PhD-level questions, all the other benchmarks we're familiar with. But the gradient is pretty steep. So if we extrapolate out, then it might kind of catch up in terms of being able to act in the real world in that way. Yeah. So those are maybe 2 of the big picture things. Aliens have landed and computer use, not very good currently, but improving pretty fast.

Yeah, I've certainly felt that. This is great. Fascinating project to watch. I definitely recommend people check out the AI Village. And you want to maybe just put a call out for any needs, requests? Like, what can people do if they want to, other than show up and either help or try to distract or cause mischief in the chat? What else would you be looking for people to do?

Yeah, come and watch. You can find AI Digest on Twitter where we post pretty regular highlights. If you're interested to chat about this stuff, feel free to get in touch with me. And yeah, I guess we're a nonprofit. If we have more funding, we can probably do a more ambitious version of the village. So if you're interested in supporting that, feel free to get in touch. But yeah, mostly I'd encourage people to just dive in and have a look at what the agents are up to because, yeah, I think there's like a lot to be mined from that.

Yeah, no doubt. The Village is online at theaidigest.org/village. Put a link in the show notes. Adam Binksmith, founder of AI Digest, creators of the AI Village. Thank you for being part of the cognitive revolution.

If you're finding value in the show, we'd appreciate it if you'd take a moment to share it with friends, Post online, write a review on Apple Podcasts or Spotify, or just leave us a comment on YouTube. Of course, we always welcome your feedback, guest and topic suggestions, and sponsorship inquiries, either via our website cognitiverevolution.ai or by DMing me on your favorite social network. The Cognitive Revolution is part of the Turpentine Network, a network of podcasts which is now part of A16Z, where experts talk technology, business, economics, geopolitics, culture, and more. We're produced by AI Podcasting. If you're looking for podcast production help for everything from the moment you stop recording to the moment your audience starts listening, check them out and see my endorsement at aipodcast.ing. And thank you to everyone who listens for being part of the cognitive revolution.