You must log in or register to comment.
An LLM is a poor computational/predictive paradigm for playing chess.
This just in: a hammer makes a poor screwdriver.
LLMs are more like a leaf blower though
Actually, a very specific model (chatgpt3.5-turbo-instruct) was pretty good at chess (around 1700 elo if i remember correctly).
I’m impressed, if that’s true! In general, an LLM’s training cost vs. an LSTM, RNN, or some other more appropriate DNN algorithm suitable for the ruleset is laughably high.
Oh yes, cost of training are ofc a great loss here, it’s not optimized at all, and it’s stuck at an average level.
Interestingly, i believe some people did research on it and found some parameters in the model that seemed to represent the state of the chess board (as in, they seem to reflect the current state of the board, and when artificially modified, the model takes modification into account in its playing). It was used by a french youtuber to show how LLMs can somehow have a kinda representation of the world. I can try to get the sources back if you’re interested.
Absolutely interested. Thank you for your time to share that.
My career path in neural networks began as a researcher for cancerous tissue object detection in medical diagnostic imaging. Now it is switched to generative models for CAD (architecture, product design, game assets, etc.). I don’t really mess about with fine-tuning LLMs.
However, I do self-host my own LLMs as code assistants. Thus, I’m only tangentially involved with the current LLM craze.
But it does interest me, nonetheless!
Here is the main blog post that i remembered : it has a follow up, a more scientific version, and uses two other articles as a basis, so you might want to dig around what they mention in the introduction.
It is indeed a quite technical discovery, and it still lacks complete and wider analysis, but it is very interesting for the fact that it kinda invalidates the common gut feeling that llms are pure lucky random.
The underlying neural network tech is the same as what the best chess AIs (AlphaZero, Leela) use. The problem is, as you said, that ChatGPT is designed specifically as an LLM so it’s been optimized strictly to write semi-coherent text first, and then any problem solving beyond that is ancillary. Which should say a lot about how inconsistent ChatGPT is at solving problems, given that it’s not actually optimized for any specific use cases.
Yes, I agree wholeheartedly with your clarification.
My career path, as I stated in a different comment in regards to neural networks, is focused on generative DNNs for CAD applications and parametric 3D modeling. Before that, I began as a researcher in cancerous tissue classification and object detection in medical diagnostic imaging.
Thus, large language models are well out of my area of expertise in terms of the architecture of their models.
However, fundamentally it boils down to the fact that the specific large language model used was designed to predict text and not necessarily solve problems/play games to “win”/“survive”.
(I admit that I’m just parroting what you stated and maybe rehashing what I stated even before that, but I like repeating and refining in simple terms to practice explaining to laymen and, dare I say, clients. It helps me feel as if I don’t come off too pompously when talking about this subject to others; forgive my tedium.)
Yeah, a lot of them hallucinate illegal moves.
Sometimes it seems like most of these AI articles are written by AIs with bad prompts.
Human journalists would hopefully do a little research. A quick search would reveal that researches have been publishing about this for over a year so there’s no need to sensationalize it. Perhaps the human journalist could have spent a little time talking about why LLMs are bad at chess and how researchers are approaching the problem.
LLMs on the other hand, are very good at producing clickbait articles with low information content.
Gotham chess has a video of making chatgpt play chess against stockfish. Spoiler: chatgpt does not do well. It plays okay for a few moves but then the moment it gets in trouble it straight up cheats. Telling it to follow the rules of chess doesn’t help.
This sort of gets to the heart of LLM-based “AI”. That one example to me really shows that there’s no actual reasoning happening inside. It’s producing answers that statistically look like answers that might be given based on that input.
For some things it even works. But calling this intelligence is dubious at best.
Because it doesn’t have any understanding of the rules of chess or even an internal model of the game state, it just has the text of chess games in its training data and can reproduce the notation, but nothing to prevent it from making illegal moves, trying to move or capture pieces that don’t exist, incorrectly declaring check/checkmate, or any number of nonsensical things.
Hallucinating 100% of the time 👌
ChatGPT versus Deepseek is hilarious. They both cheat like crazy and then one side jedi mind tricks the winner into losing.
So they are both masters of troll chess then?
See: King of the Bridge
It plays okay for a few moves but then the moment it gets in trouble it straight up cheats.
Lol. More comparisons to how AI is currently like a young child.
I think the biggest problem is it’s very low ability to “test time adaptability”. Even when combined with a reasonning model outputting into its context, the weights do not learn out of the immediate context.
I think the solution might be to train a LoRa overlay on the fly against the weights and run inference with that AND the unmodified weights and then have an overseer model self evaluate and recompose the raw outputs.
Like humans are way better at answering stuff when it’s a collaboration of more than one person. I suspect the same is true of LLMs.
Like humans are way better at answering stuff when it’s a collaboration of more than one person. I suspect the same is true of LLMs.
It is.
It’s really common for non-language implementations of neural networks. If you have an NN that’s right some percentage of the time, you can often run it through a bunch of copies of the NNs and take the average and that average is correct a higher percentage of the time.
Aider is an open source AI coding assistant that lets you use one model to plan the coding and a second one to do the actual coding. It works better than doing it in a single pass, even if you assign the the same model to planing and coding.
In this case it’s not even bad prompts, it’s a problem domain ChatGPT wasn’t designed to be good at. It’s like saying modern medicine is clearly bullshit because a doctor loses a basketball game.
I imagine the “author” did something like, “Search http://google.scholar.com/ find a publication where AI failed at something and write a paragraph about it.”
It’s not even as bad as the article claims.
Atari isn’t great at chess. https://chess.stackexchange.com/questions/24952/how-strong-is-each-level-of-atari-2600s-video-chess
Random LLMs were nearly as good 2 years ago. https://lmsys.org/blog/2023-05-03-arena/
LLMs that are actually trained for chess have done much better. https://arxiv.org/abs/2501.17186Wouldn’t surprise me if an LLM trained on records of chess moves made good chess moves. I just wouldn’t expect the deployed version of ChatGPT to generate coherent chess moves based on the general text it’s been trained on.
I wouldn’t either but that’s exactly what lmsys.org found.
That blog post had ratings between 858 and 1169. Those are slightly higher than the average rating of human users on popular chess sites. Their latest leaderboard shows them doing even better.
https://lmarena.ai/leaderboard has one of the Gemini models with a rating of 1470. That’s pretty good.
Ah, you used logic. That’s the issue. They don’t do that.
2025 Mazda MX-5 Miata ‘got absolutely wrecked’ by Inflatable Boat in beginner’s boat racing match — Mazda’s newest model bamboozled by 1930s technology.
Using an LLM as a chess engine is like using a power tool as a table leg. Pretty funny honestly, but it’s obviously not going to be good at it, at least not without scaffolding.
is like using a power tool as a table leg.
Then again, our corporate lords and masters are trying to replace all manner of skilled workers with those same LLM “AI” tools.
And clearly that will backfire on them and they’ll eventually scramble to find people with the needed skills, but in the meantime tons of people will have lost their source of income.
If you believe LLMs are not good at anything then there should be relatively little to worry about in the long-term, but I am more concerned.
It’s not obvious to me that it will backfire for them, because I believe LLMs are good at some things (that is, when they are used correctly, for the correct tasks). Currently they’re being applied to far more use cases than they are likely to be good at – either because they’re overhyped or our corporate lords and masters are just experimenting to find out what they’re good at and what not. Some of these cases will be like chess, but others will be like code*.
(* not saying LLMs are good at code in general, but for some coding applications I believe they are vastly more efficient than humans, even if a human expert can currently write higher-quality less-buggy code.)
I believe LLMs are good at some things
The problem is that they’re being used for all the things, including a large number of tasks that thwy are not well suited to.
yeah, we agree on this point. In the short term it’s a disaster. In the long-term, assuming AI’s capabilities don’t continue to improve at the rate they have been, our corporate overlords will only replace people for whom it’s actually worth it to them to replace with AI.
All these comments asking “why don’t they just have chatgpt go and look up the correct answer”.
That’s not how it works, you buffoons, it trains off of datasets long before it releases. It doesn’t think. It doesn’t learn after release, it won’t remember things you try to teach it.
Really lowering my faith in humanity when even the AI skeptics don’t understand that it generates statistical representations of an answer based on answers given in the past.
If you don’t play chess, the Atari is probably going to beat you as well.
LLMs are only good at things to the extent that they have been well-trained in the relevant areas. Not just learning to predict text string sequences, but reinforcement learning after that, where a human or some other agent says “this answer is better than that one” enough times in enough of the right contexts. It mimics the way humans learn, which is through repeated and diverse exposure.
If they set up a system to train it against some chess program, or (much simpler) simply gave it a tool call, it would do much better. Tool calling already exists and would be by far the easiest way.
It could also be instructed to write a chess solver program and then run it, at which point it would be on par with the Atari, but it wouldn’t compete well with a serious chess solver.
Can i fistfight ChatGPT next? I bet I could kick its ass, too :p
this is because an LLM is not made for playing chess
Is anyone actually surprised at that?
So, it fares as well as the average schmuck, proving it is human
/s
Llms useless confirmed once again
Isn’t the Atari just a game console, not a chess engine?
Like, Wikipedia doesn’t mention anything about the Atari 2600 having a built-in chess engine.
If they were willing to run a chess game on the Atari 2600, why did they not apply the same to ChatGPT? There are custom GPTs which claim to use a stockfish API or play at a similar level.
Like this, it’s just unfair. Both platforms are not designed to deal with the task by themselves, but one of them is given the necessary tooling, the other one isn’t. No matter what you think of ChatGPT, that’s not a fair comparison.
Edit: Given the existing replies and downvotes, I think this comment is being misunderstood. I would like to try clarifying again what I meant here.
First of all, I’d like to ask if this article is satire. That’s the only way I can understand the replies I’ve gotten that critized me on grounds of the marketing aspect of LLMs (when the article never brings up that topic itself, nor did I). Like, if this article is just some tongue in cheek type thing about holding LLMs to the standards they’re advertised at, I can understand both the article and the replies I’ve gotten. But the article never suggests so itself. So my assumption when writing my comment was that this is not the case and it is serious.
The Atari is hardware. It can’t play chess on its own. To be able to, you need a game for it which is inserted. Then the Atari can interface with the cartridge and play the game.
ChatGPT is an LLM. Guess what, it also can’t play chess on its own. It also needs to interface with a third party tool that enables it to play chess.
Neither the Atari nor ChatGPT can directly, on their own, play chess. This was my core point.
I merely pointed out that it’s unfair that one party in this comparison is given the tool it needs (the cartridge), but the other party isn’t. Unless this is satire, I don’t see how marketing plays a role here at all.
GPTs which claim to use a stockfish API
Then the actual chess isn’t LLM. If you are going stockfish, then the LLM doesn’t add anything, stockfish is doing everything.
The whole point is the marketing rage is that LLMs can do all kinds of stuff, doubling down on this with the branding of some approaches as “reasoning” models, which are roughly “similar to ‘pre-reasoning’, but forcing use of more tokens on disposable intermediate generation steps”. With this facet of LLM marketing, the promise would be that the LLM can “reason” itself through a chess game without particular enablement. In practice, people trying to feed in gobs of chess data to an LLM end up with an LLM that doesn’t even comply to the rules of the game, let alone provide reasonable competitive responses to an oppone.
Then the actual chess isn’t LLM.
And neither did the Atari 2600 win against ChatGPT. Whatever game they ran on it did.
That’s my point here. The fact that neither Atari 2600 nor ChatGPT are capable of playing chess on their own. They can only do so if you provide them with the necessary tools. Which applies to both of them. Yet only one of them was given those tools here.
Fine, a chess engine that is capable of running with affordable even for the time 1970s electronics will best what marketing folks would have you think is an arbitrarily capable “reasoning” model running on top of the line 2025 hardware.
You can split hairs about “well actually, the 2600 is hardware and a chess engine is the software” but everyone gets the point.
As to assertions that no one should expect an LLM to be a chess engine, well tell that to the industry that is asserting the LLMs are now “reasoning” and provides a basis to replace most of the labor pool. We need stories like this to calibrate expectations in a way common people can understand…
The Atari 2600 is just hardware. The software came on plug-in cartridges. Video Chess was released for it in 1979.
Did the author thinks ChatGPT is in fact an AGI? It’s a chatbot. Why would it be good at chess? It’s like saying an Atari 2600 running a dedicated chess program can beat Google Maps at chess.
OpenAI has been talking about AGI for years, implying that they are getting closer to it with their products.
https://openai.com/index/planning-for-agi-and-beyond/
https://openai.com/index/elon-musk-wanted-an-openai-for-profit/
Not to even mention all the hype created by the techbros around it.
Hey I didn’t say anywhere that corporations don’t lie to promote their product did I?
You’re not wrong, but keep in mind ChatGPT advocates, including the company itself are referring to it as AI, including in marketing. They’re saying it’s a complete, self-learning, constantly-evolving Artificial Intelligence that has been improving itself since release… And it loses to a 4KB video game program from 1979 that can only “think” 2 moves ahead.
That’s totally fair, the company is obviously lying, excuse me “marketing”, to promote their product, that’s absolutely true.
AI including ChatGPT is being marketed as super awesome at everything, which is why that and similar AI is being forced into absolutely everything and being sold as a replacement for people.
Something marketed as AGI should be treated as AGI when proving it isn’t AGI.
Not to help the AI companies, but why don’t they program them to look up math programs and outsource chess to other programs when they’re asked for that stuff? It’s obvious they’re shit at it, why do they answer anyway? It’s because they’re programmed by know-it-all programmers, isn’t it.
why don’t they program them to look up math programs and outsource chess to other programs when they’re asked for that stuff?
Because the AI doesn’t know what it’s being asked, it’s just a algorithm guessing what the next word in a reply is. It has no understanding of what the words mean.
“Why doesn’t the man in the Chinese room just use a calculator for math questions?”
Because they’re fucking terrible at designing tools to solve problems, they are obviously less and less good at pretending this is an omnitool that can do everything with perfect coherency (and if it isn’t working right it’s because you’re not believing or paying hard enough)
Or they keep telling you that you just have to wait it out. It’s going to get better and better!
why don’t they program them
AI models aren’t programmed traditionally. They’re generated by machine learning. Essentially the model is given test prompts and then given a rating on its answer. The model’s calculations will be adjusted so that its answer to the test prompt will be closer to the expected answer. You repeat this a few billion times with a few billion prompts and you will have generated a model that scores very high on all test prompts.
Then someone asks it how many R’s are in strawberry and it gets the wrong answer. The only way to fix this is to add that as a test prompt and redo the machine learning process which takes an enormous amount of time and computational power each time it’s done, only for people to once again quickly find some kind of prompt it doesn’t answer well.
There are already AI models that play chess incredibly well. Using machine learning to solve a complexe problem isn’t the issue. It’s trying to get one model to be good at absolutely everything.
…or a simple counter to count the r in strawberry. Because that’s more difficult than one might think and they are starting to do this now.
Because the LLMs are now being used to vibe code themselves.
They are starting to do this. Most new models support function calling and can generate code to come up with math answers etc
This is where MCP comes in. It’s a protocol for LLMs to call standard tools. Basically the LLM would figure out the tool to use from the context, then figure out the order of parameters from those the MCP server says is available, send the JSON, and parse the response.
If you pay for chatgpt you can connect it with wolfrenalpha and it’s relays the maths to it
I don’t pay for ChatGPT and just used the Wolfram GPT. They made the custom GPTs non-paid at some point.
I think they’re trying to do that. But AI can still fail at that lol
From a technology standpoint, nothing is stopping them. From a business standpoint: hubris.
To put time and effort into creating traditional logic based algorithms to compensate for this generic math model would be to admit what mathematicians and scientists have known for centuries. That models are good at finding patterns but they do not explain why a relationship exists (if it exists at all). The technology is fundamentally flawed for the use cases that OpenAI is trying to claim it can be used in, and programming around it would be to acknowledge that.
why don’t they program them to look up math programs and outsource chess to other programs when they’re asked for that stuff?
They will, when it makes sense for what the AI is designed to do. For example, ChatGPT can outsource image generation to an AI dedicated to that. It also used to calculate math using python for me, but that doesn’t seem to happen anymore, probably due to security issues with letting the AI run arbitrary python code.
ChatGPT however was not designed to play chess, so I don’t see why OpenAI should invest resources into connecting it to a chess API.
I think especially since adding custom GPTs, adding this kind of stuff has become kind of unnecessary for base ChatGPT. If you want a chess engine, get a GPT which implements a Stockfish API (there seem to be several GPTs that do). For math, get the Wolfram GPT which uses Wolfram Alpha’s API, or a different powerful math GPT.
I don’t think ai is being marketed as awesome at everything. It’s got obvious flaws. Right now its not good for stuff like chess, probably not even tic tac toe. It’s a language model, its hard for it to calculate the playing field. But ai is in development, it might not need much to start playing chess.
What the tech is being marketed as and what it’s capable of are not the same, and likely never will be. In fact all things are very rarely marketed how they truly behave, intentionally.
Everyone is still trying to figure out what these Large Reasoning Models and Large Language Models are even capable of; Apple, one of the largest companies in the world just released a white paper this past week describing the “illusion of reasoning”. If it takes a scientific paper to understand what these models are and are not capable of, I assure you they’ll be selling snake oil for years after we fully understand every nuance of their capabilities.
TL;DR Rich folks want them to be everything, so they’ll be sold as capable of everything until we repeatedly refute they are able to do so.
I think in many cases people intentionally or unintentionally disregard the time component here. Ai is in development. I think what is being marketed here, just like in the stock market, is a piece of the future. I don’t expect the models I use to be perfect and not make mistakes, so I use them accordingly. They are useful for what I use them for and I wouldn’t use them for chess. I don’t expect that laundry detergent to be just as perfect in the commercial either.
Marketing does not mean functionality. AI is absolutely being sold to the public and enterprises as something that can solve everything. Obviously it can’t, but it’s being sold that way. I would bet the average person would be surprised by this headline solely on what they’ve heard about the capabilities of AI.
I don’t think anyone is so stupid to believe current ai can solve everything.
And honestly, I didn’t see any marketing material that would claim that.
You are both completely over estimating the intelligence level of “anyone” and not living in the same AI marketed universe as the rest of us. People are stupid. Really stupid.
I don’t understand why this is so important, marketing is all about exaggerating, why expect something different here.
It’s not important. You said AI isn’t being marketed to be able to do everything. I said yes it is. That’s it.
The Zoom CEO, that is the video calling software, wanted to train AIs on your work emails and chat messages to create AI personalities you could send to the meetings you’re paid to sit through while you drink Corona on the beach and receive a “summary” later.
The Zoom CEO, that is the video calling software, seems like a pretty stupid guy?
Yeah. Yeah, he really does. Really… fuckin’… dumb.
Same genius who forced all his own employees back into the office. An incomprehensibly stupid maneuver by an organization that literally owes its success to people working from home.
Really then why are they cramming AI into every app and every device and replacing jobs with it and claiming they’re saving so much time and money and they’re the best now the hardest working most efficient company and this is the future and they have a director of AI vision that’s right a director of AI vision a true visionary to lead us into the promised land where we will make money automatically please bro just let this be the automatic money cheat oh god I’m about to
Those are two different things.
-
they are craming ai everywhere because nobody wants to miss the boat and because it plays well in the stock market.
-
the people claiming it’s awesome and that they are doing I don’t know what with it, replacing people are mostly influencers and a few deluded people.
Ai can help people in many different roles today, so it makes sense to use it. Even in roles that is not particularly useful, it makes sense to prepare for when it is.
it makes sense to prepare for when it is.
Pfft, okay.
-
Most people do. It’s just called AI in the media everywhere and marketing works. I think online folks forget that something as simple as getting a Lemmy account by yourself puts you into the top quintile of tech literacy.
Yet even on Lemmy people can’t seem to make sense of these terms and are saying things like “LLM’s are not AI”
deleted by creator
Google Maps doesn’t pretend to be good at chess. ChatGPT does.
A toddler can pretend to be good at chess but anybody with reasonable expectations knows that they are not.
Plot twist: the toddler has a multi-year marketing push worth tens if not hundreds of millions, which convinced a lot of people who don’t know the first thing about chess that it really is very impressive, and all those chess-types are just jealous.
Have you tried feeding the toddler gallons of baby-food? Maybe then it can play chess
They’ve been feeding the toddler everybody else’s baby food and claiming they have the right to.
“If we have to ask every time before stealing a little baby food, our morbidly obese toddler cannot survive”
well so much hype has been generated around chatgpt being close to AGI that now it makes sense to ask questions like “can chatgpt prove the Riemann hypothesis”
Even the models that pretend to be AGI are not. It’s been proven.
I agree with your general statement, but in theory since all ChatGPT does is regurgitate information back and a lot of chess is memorization of historical games and types, it might actually perform well. No, it can’t think, but it can remember everything so at some point that might tip the results in it’s favor.
Regurgitating an impression of, not regurgitating verbatim, that’s the problem here.
Chess is 100% deterministic, so it falls flat.
I’m guessing it’s not even hard to get it to “confidently” violate the rules.
I mean it may be possible but the complexity would be so many orders of magnitude greater. It’d be like learning chess by just memorizing all the moves great players made but without any context or understanding of the underlying strategy.
I think that’s generally the point is most people thing chat GPT is this sentient thing that knows everything and… no.
Do they though? No one I talked to, not my coworkers that use it for work, not my friends, not my 72 year old mother think they are sentient.
Okay I maybe exaggerated a bit, but a lot of people think it actually knows things, or is actually smart. Which… it’s not… at all. It’s just pattern recognition. Which was I assume the point of showing it can’t even beat the goddamn Atari because it cannot think or reason, it’s all just copy pasta and pattern recognition.
Articles like this are good because it exposes the flaws with the ai and that it can’t be trusted with complex multi step tasks.
Helps people see that think AI is close to a human that its not and its missing critical functionality
The problem is though that this perpetuates the idea that ChatGPT is actually an AI.
People already think chatGPT is a general AI. We need more articles like this showing is ineffectiveness at being intelligent. Besides it helps find a limitations of this technology so that we can hopefully use it to argue against every single place
In all fairness. Machine learning in chess engines is actually pretty strong.
AlphaZero was developed by the artificial intelligence and research company DeepMind, which was acquired by Google. It is a computer program that reached a virtually unthinkable level of play using only reinforcement learning and self-play in order to train its neural networks. In other words, it was only given the rules of the game and then played against itself many millions of times (44 million games in the first nine hours, according to DeepMind).
Sure, but machine learning like that is very different to how LLMs are trained and their output.
Oh absolutely you can apply machine learning to game strategy. But you can’t expect a generalized chatbot to do well at strategic decision making for a specific game.
I like referring to LLMs as VI (Virtual Intelligence from Mass Effect) since they merely give the impression of intelligence but are little more than search engines. In the end all one is doing is displaying expected results based on a popularity algorithm. However they do this inconsistently due to bad data in and limited caching.
I mean, open AI seem to forget it isn’t.
Tbf, the article should probably mention the fact that machine learning programs designed to play chess blow everything else out of the water.
Machine learning has existed for many years, now. The issue is with these funding-hungry new companies taking their LLMs, repackaging them as “AI” and attributing every ML win ever to “AI”.
ML programs designed and trained specifically to identify tumors in medical imaging have become good diagnostic tools. But if you read in news that “AI helps cure cancer”, it makes it sound like it was a lone researcher who spent a few minutes engineering the right prompt for Copilot.
Yes a specifically-designed and finely tuned ML program can now beat the best human chess player, but calling it “AI” and bundling it together with the latest Gemini or Claude iteration’s “reasoning capabilities” is intentionally misleading. That’s why articles like this one are needed. ML is a useful tool but far from the “super-human general intelligence” that is meant to replace half of human workers by the power of wishful prompting
Yeah its like judging how great a fish is at climbing a tree. But it does show that it’s not real intelligence or reasoning
Don’t call my fish stupid.
Well, can it climb trees?
I forgot which airline it is but one of the onboard games in the back of a headrest TV was a game called “Beginners Chess” which was notoriously difficult to beat so it was tested against other chess engines and it ranked in like the top five most powerful chess engines ever
It does
It does not. Where?
