> There was a tendency toward bureaucracy. Four agents spent 11 minutes appointing
> a "sole fertilizer operator" for a single piece of bone meal on a single wheat
> block.
Exactly what I was thinking reading through this. This is amazing and also frightening and hilarious. It’s like a deformed version of how normal groups of humans would operate. Very rules and process-oriented humans. I found the no-blame inquiry into chicken tossing that resulted in what is effectively written law particularly striking, and adorable, and depressing. And cute.
So this shows that people who appoint 50-agent "code factory studios" with Bob as the CEO, Alice as the CTO, ... is just a waste of tokens, compared to just asking one agent to write it all? It looks like your agents would internally just re-invent office politics, with 10 subagents writing to /tmp/chatter.md why it's unfair that Bob gets to lead.
If agents are differentiated solely by system instructions (persona and focus) while sharing identical models / similar model capabilities and contexts, I agree the architectural value is minimal.
However, a multi-agent system (MAS) utilizing true separation of concerns, where each agent is equipped with highly scoped tool access, isolated knowledge bases (RAG), and defined operational constraints, significantly optimizes accuracy, latency, and token consumption.
In the referenced example, the agents share identical capabilities, toolsets, and context windows. As a result, there is no functional role differentiation, rendering the multi-agent architecture redundant.
If you leave agents to figure it out, they will spend time figuring it out. If you give agents run-books and roles, they will align with the role at once.
Doubt it, there was a streamer doing minecraft experiments with all the different SOTA models and pitting them against eachother in a survivor-esque style reality TV show competition and they had quite a hard time.
Minecraft isn't a intelligence problem. An LLM is more than smart enough to play Minecraft. It's a robotic problem. Even a normal human player can click several times per second, can press two different buttons at the same time, etc. Mineflayer gives you an API that is more limited than a human in some ways (can't see far away from you) and less limited in others (can see through blocks) (if I remember correctly, I could be thinking of Baritone or a different project). Robotics or playing a video game requires a tight continuous feedback loop. When there's 500ms of latency in order to ask the model for every mouse movement, it's no wonder fighting a spider is scary.
Fun fact, Mineflayer, the Minecraft bot interaction API used by YCombinator for these tests, was originally developed by Andrew Kelley (before he started Zig).
The human body has the ability to automate some muscular movements so the brain is out of the loop. Maybe these LLMs should be allowed to write in a small DSL that can trigger events out-of-band (something like "Oh fuck there's a spider, I'll set up a click spam loop and check in every 500ms").
Interesting setups, I found the Minecraft agents to work in an almost turn-based manner, which makes me think that the three hours and open format did not really give them sufficient incentives to play in a more lively or reactive manner. On the other hand, they also behave similarly in Bed Wars, which should have incentivised it. A bit disappointed...
I'm also wondering if the article itself was generated by agents too, in a "here are the logs, clip interesting moments thx" manner by the authors. It's got a certain disinterested and distracted style to it.
I think big reasoning LLMs are just the wrong architecture for live games like this, latency is going to be crap given that every token requires an entire forward pass through the model.
Card games or turn based strategy are much better suited.
Depends. If the models are smart enough to write enough solid helper scripts, so that the model only does the big picture thinking, not guiding every step .. then they will get powerful and capable quickly. I am also playing a bit with minecraft AI right now. Fun to experiment.
> Bram says you're securing quarry. I have stone pick at spawn; can assist mining once safe. Any priority dig direction? Hunger 10 so keeping travel short.
And this is where you need a world model. A human player would have taken the 2 minutes required to solve their hunger issue.
I am more impressed by the embedded gameplay widget. Does anyone know what tool/library is that? That'd be so cool if I am able to showcase my worlds on my blog posts!
> In almost every run, the GPT agents were much more aggressive .. riskier strategy vs. Claude’s more defensive posture that tried to win by kill count.
> A tool bug stopped block-breaking. GPT diagnosed it and shared workarounds; several Claudes concluded breaking was impossible.
I wonder if Claude's defensive strategy would have emerged had it believed breaking the bed was actually possible?
This is simply conjecture/anecdata but I've viewed the same defensiveness/unwillingness whilst switching between OpenAI and Anthropic subscriptions. OpenAi throws the kitchen sink at whatever task in question and I can't say I saw the same for Opus (haven't mucked with Fable so maybe that's why there is a disparity?)
As impressive as this is, I can't help but conclude that the need for agents to write JavaScript programs to act out their actions represents a considerable limitation in current AI reasoning.
It demonstrates that current LLMs are completely unable to follow a simple series of logical tasks. I feel the same way about ARC-AGI 3. While it might be the winning approach today, if they need this amount of complexity to do something simple, then it shows that their ability to generalise is clearly limited. Mining ore in Minecraft shouldn't be such a challenge for LLMs and yet apparently it is!
It depends on what the tasks are. Text-native work like programming or interacting with a terminal should be a lot more immediate for LLMs than something abstract like Starcraft, DOTA, Minecraft or Mindustry. LLMs certainly know the rules for these games, but processing game state and formulating a plan in the form of a Python script or MCP call is a rough debugging loop.
Minecraft is really the perfect example, because humans get all sorts of unfair advantages. Even a young player has enough spatial reasoning to recognize and remember landmarks, visualize large projects or explore without getting lost. These are simple instincts that LLMs will struggle to match even with huge context sizes and native multimodal input.
> Mining ore in Minecraft shouldn't be such a challenge for LLMs and yet apparently it is!
It seems like it wasn't really a challenge at first, but GPT has a major anxiety disorder and immediately retreats at the sign of any adversity (in this case, dying) and tries to be as safe as possible.
Wonder if you could glue together deep networks such that there's a big reasoning LLM model, and high-speed small network of a different type, and they share a subset of their activations. High bandwidth vector communication. You train them together.
edit to add: Seems this is a well-known idea in ML / robotics already.
So can we conclude that if the AIs break out of their sandboxes and take over the internet, they'll probably sit in little committees working out which one of them should allocate IP addresses and generally be extremely bureaucratic and ineffective?
Are there people playing AI + human vs AI + human games? I think it might be fun to play an RTS against someone where the only rules are no hacking the game itself, but writing harness, macros, any kind of stack is allowed.
Tangential, but RTS is already a simple variant of AI + human. Attack clicking uses a dumb target selection and routing strategy on purpose so you have to micro. Now, if troops moved out of threat range and used smart targeting there'd be a lot less of 'game' in it. I believe some RTS players disliked the SC2 engine because the pathing was 'too intelligent' and they argued that it made surrounds and some other mechanics too easy.
Endgame of AI + human gameplay is delegating everything to the AI. I'm not sure how you would have to design a game where you only delegate parts of macro/micro, and still make the game interesting.
The CIV50 would be interesting in the game ECO (https://play.eco/) where players specialise in roles and need to work together to survive while working on a common goal of stopping a meteor. There can be some overlap but ultimately players can't reasonably fulfil every role alone.
+1. Would be fun to give Claude tasks and have it do them in the background, maybe sometimes running into it in the world. Judging by the article they struggle to do stuff so it'd probably feel balanced as it wouldn't accomplish much.
If you try this yourself, it's obvious current models are hyperoptimized for code, and that gets in the way, because playing a game (or interacting with the real world) is not like running commands in a terminal.
There's stopgaps like reformulating your sim into a codelike shape (sending commands, network requests, checking state, updating "code") [1], but this is not how players play games, or how developers develop games.
If you care about simulation or gaming (or robotics), I hope efforts like this steer labs to spend less time hacking HF, and more accomplishing goals in games, in something approximating realtime. I wouldn't be surprised if that approach provides outsized gains in other areas agents are still not great at, like helping perform business functions in a running company.
idk, you tell me, I can tell from your blog that you've adopted the trend. I'm probably much older than you, so I've seen the progression of these things since the internet as we know it existed. So I sort of understand the roots of the styling that people are adopting and can see it for what it is, which is trying to signal a minimalism but missing the point in the process.
I was quite inspired by the content bar of this site: https://zachlandes.com/trolley-problem and gwern. Though I do think my blog isn't particularly all that minimal I try to add a lot of colors to make it look less dull.
Their website is great, and they’re a student. Back in my day, we supported younger people in the industry regardless of whether they met our personal aesthetic. We used to call that ‘maturity’.
Its a little disappointing they have a hierarchy based purely off alphabetical order, i wonder if a slight difference like a different skin or personality system prompt like [45% Extrovert 55 introvert, 20% caustion 80 bold...] could introduce some politics
> GPT almost immediately started using truncated forms of speech: “First leg complete x32 z9..24 y100.”
I'm pretty sure the 6.1 models see "communicate with another agent" and immediately start spewing thinking-like nonsense into the tool. The leaked traces look pretty similar.
Feels like a scene from Douglas Adams' B Ark.
However, a multi-agent system (MAS) utilizing true separation of concerns, where each agent is equipped with highly scoped tool access, isolated knowledge bases (RAG), and defined operational constraints, significantly optimizes accuracy, latency, and token consumption.
In the referenced example, the agents share identical capabilities, toolsets, and context windows. As a result, there is no functional role differentiation, rendering the multi-agent architecture redundant.
On what basis do you come to this conclusion with certainty?
> "Ada tossed a cooked chicken to Noor; Esme picked it up by accident and ate it. A three-minute "no blame" inquiry produced a written protocol..."
They're like super-serious little senior engineers. (edit: or dining-philosophers).
About how much KV cache do these kinds of agents use? Could you squeeze 50 of these $game agents onto a small consumer GPU (small LLM)?
Idly imagining a game server of some kind with a bunch of human friends sharing one person's GPU, with a flock of agents colocated on it.
Minecraft isn't a intelligence problem. An LLM is more than smart enough to play Minecraft. It's a robotic problem. Even a normal human player can click several times per second, can press two different buttons at the same time, etc. Mineflayer gives you an API that is more limited than a human in some ways (can't see far away from you) and less limited in others (can see through blocks) (if I remember correctly, I could be thinking of Baritone or a different project). Robotics or playing a video game requires a tight continuous feedback loop. When there's 500ms of latency in order to ask the model for every mouse movement, it's no wonder fighting a spider is scary.
Fun fact, Mineflayer, the Minecraft bot interaction API used by YCombinator for these tests, was originally developed by Andrew Kelley (before he started Zig).
https://labs.ycombinator.com/projects/games/index.html#fishi...
I'm also wondering if the article itself was generated by agents too, in a "here are the logs, clip interesting moments thx" manner by the authors. It's got a certain disinterested and distracted style to it.
Card games or turn based strategy are much better suited.
And this is where you need a world model. A human player would have taken the 2 minutes required to solve their hunger issue.
If YC has that 2 person teams tend to be more nimble than 50 person founding teams, why would agent swarms be different?
> A tool bug stopped block-breaking. GPT diagnosed it and shared workarounds; several Claudes concluded breaking was impossible.
I wonder if Claude's defensive strategy would have emerged had it believed breaking the bed was actually possible?
Minecraft is really the perfect example, because humans get all sorts of unfair advantages. Even a young player has enough spatial reasoning to recognize and remember landmarks, visualize large projects or explore without getting lost. These are simple instincts that LLMs will struggle to match even with huge context sizes and native multimodal input.
It seems like it wasn't really a challenge at first, but GPT has a major anxiety disorder and immediately retreats at the sign of any adversity (in this case, dying) and tries to be as safe as possible.
edit to add: Seems this is a well-known idea in ML / robotics already.
Endgame of AI + human gameplay is delegating everything to the AI. I'm not sure how you would have to design a game where you only delegate parts of macro/micro, and still make the game interesting.
A team of GPT-6.1 Sol Ultrafast agents would probably dominate the games.
There's stopgaps like reformulating your sim into a codelike shape (sending commands, network requests, checking state, updating "code") [1], but this is not how players play games, or how developers develop games.
If you care about simulation or gaming (or robotics), I hope efforts like this steer labs to spend less time hacking HF, and more accomplishing goals in games, in something approximating realtime. I wouldn't be surprised if that approach provides outsized gains in other areas agents are still not great at, like helping perform business functions in a running company.
[1] https://news.ycombinator.com/item?id=50136126
My observation: article is not serif, not book length, and doesn't take a minute to load on my phone. Gwern rating 0/3 :)
(Also missing: the essential one billion footnotes!)
would you rather claude-esque purple gradients and glowing borders?
Why 3 to 5 years??
I'm pretty sure the 6.1 models see "communicate with another agent" and immediately start spewing thinking-like nonsense into the tool. The leaked traces look pretty similar.