One comment about the source (chess.com): it has a conflict of interest in this topic and has a clear bias in what it does and doesn't say. They omit that the cheater in question is a prominent member of their own chess platform, who has been scoring first-place in high-profile sitewide tournaments [0]. IMO the most interesting question here, which this source would prefer to elide over, is whether or not this over-the-board (meatspace) cheater has also been cheating for profit on the internet platforms—something that's technically much easier to do than in a professional, arbitrated tournament setting.
This is sort of an existential question for their business. Chess.com currently operates one of the two major online chess platforms. Their top public-image controversy, going back years, is whether or not their anti-computer cheating methods are effective and accurate. (This question has even made it into the Wall Street Journal [1]). Their system is a "security by obscurity" type, which both relies on cheaters not knowing how it works, and asks paying customers to trust the company's claims, without evidence, that their competitive platform is fair and fundamentally sound.
I think this revelation is going to (rightly) enflame that debate—though you won't read much about it on chess.com's own blog.
GM Denis Kadric (currently rated 309th worldwide) was just banned by Chess.com for allegedly cheating, and has no ability to appeal or see evidence used in the decision.
Not only is the website the de facto platform for online play, some tournaments also use it for in-person games. The site has essentially become a chess authority organization of its own right. Its staff operate as judge, jury, and executioner behind closed doors. While I’m sure the majority of banned accounts were blatantly cheating, their team has apparently taken input from ChatGPT to run “simulations” to determine cheating probabilities.
I don't play chess and have never used chess.com, what's to stop someone playing on their PC and using an app on their phone to prompt their next move? I guess their anti cheat can't be like video game DRM but how do you declare someone's moved to be cheating without proving they're using a 3rd party?
It's a set of statistical methods as far as I can gather, that look at things like timing tells as well as accuracy compared to engine moves and so on. Obviously it's going to be imperfect.
In the same way mining a bitcoin proves that you've computed a large number of SHA-256 hashes, playing certain sequences of chess moves is attestation that you've run a large number of chess calculations—that you've solved a certain large system of constraints that has no real solution other than brute force. Naive cheating (that thing you describe) is obvious, because the cheater is running 6 orders of magnitude more computational power than a human and is proving it to you.
For me I would say things started to get strange at move six. He played his move and left the playing hall for more than 10 minutes and this was repeated many times in the next moves. I thought he had some stomach problems! But at some point I decided to go out to see where he is and he was standing outside the toilet room and when he saw me he went back to the playing hall. And then at some point in the game he stopped going out till the game was finished.
Multiple extended length “bathroom” breaks and he thought nobody would be suspicious? He probably could have just consulted for one move and nobody would have caught wise.
some basic level of formal culture in chess went
missing.
Chess is a battle in a war,fought
with the mind,and conforming to the ancient rules
of engagement and decorum.
Certainly in the recent past,the outcome of a match between the representitives of adversarial nations was treated as proof of the intelectual
and educational prowess of the winning side.
What we are watching is the inevitable consiquence of comodifying everything,and the ho hummyness of
the incident bieng disscused is the sound of the door closing.
This guy is 2656 and the world number 69. You might think, meh, so what who's heard of the world number 69 at, say, tennis. But if you are a serious chess player you understand that is an astronomically high level that is completely out of reach for 99.9% of people even if you love the game and devote many years of effort to it. (No doubt the same sentiment would occur to a serious tennis player in the tennis analogy).
I have two points really A) As a chess player and a programmer, I think many a programmer fancies themselves as being a rock star performer but if there was an equivalent of Elo ratings for programmers it would likely be a very sobering experience for almost everybody [I think I am a highly skilled programmer, but my chess rating is 1867].
And B) It is consequently a real personal tragedy for this guy that he has foolishly flushed his career and reputation at age 22. What a waste.
For reference, chess skill has a function indicating likelihood to beat a player of another rank (I don't recall the ratios). This user's chess skill dwarfs mine (1400s), might be higher than anyone I've ever played with in person. I can stomp all my acquaintances easily, but this person might never lose a game to me, no matter how many we play. There's something deeply gratifying about the formalization of chess skill like this imo, it keeps everyone humble. It's good to meet other players, cheers.
That said, I'm only surprised that such a high-ranking player was caught so easily with cheating. I've been expecting for many years that the problem of catching cheating in chess is complicated by motivated cheaters being necessarily intellectual, therefore hiding it well. We may never know our best cheaters, at least during their professional careers, and it's a little disheartening knowing even top100 players can overtly toss away their prospects this way. It would be a surprise for anyone this high to not be invested in chess in multiple financial avenues, so this really is like shooting yourself in the foot.
One comment about the source (chess.com): it has a conflict of interest in this topic and has a clear bias in what it does and doesn't say. They omit that the cheater in question is a prominent member of their own chess platform, who has been scoring first-place in high-profile sitewide tournaments [0]. IMO the most interesting question here, which this source would prefer to elide over, is whether or not this over-the-board (meatspace) cheater has also been cheating for profit on the internet platforms—something that's technically much easier to do than in a professional, arbitrated tournament setting.
This is sort of an existential question for their business. Chess.com currently operates one of the two major online chess platforms. Their top public-image controversy, going back years, is whether or not their anti-computer cheating methods are effective and accurate. (This question has even made it into the Wall Street Journal [1]). Their system is a "security by obscurity" type, which both relies on cheaters not knowing how it works, and asks paying customers to trust the company's claims, without evidence, that their competitive platform is fair and fundamentally sound.
I think this revelation is going to (rightly) enflame that debate—though you won't read much about it on chess.com's own blog.
[0] https://www.chess.com/news/view/shevchenko-svane-win-titled-... ("GMs Kirill Shevchenko and Frederik Svane are your winners for the October 8 editions of Titled Tuesday")
[1] https://news.ycombinator.com/item?id=33085649 ("Chess Investigation Finds U.S. Grandmaster ‘Likely Cheated’ More Than 100 Times (wsj.com)" (2022))
GM Denis Kadric (currently rated 309th worldwide) was just banned by Chess.com for allegedly cheating, and has no ability to appeal or see evidence used in the decision.
https://old.reddit.com/r/chess/comments/1fz4snq/gm_denis_kad...
Not only is the website the de facto platform for online play, some tournaments also use it for in-person games. The site has essentially become a chess authority organization of its own right. Its staff operate as judge, jury, and executioner behind closed doors. While I’m sure the majority of banned accounts were blatantly cheating, their team has apparently taken input from ChatGPT to run “simulations” to determine cheating probabilities.
https://old.reddit.com/r/ChatGPT/comments/1873ocq/chesscom_c...
I don't play chess and have never used chess.com, what's to stop someone playing on their PC and using an app on their phone to prompt their next move? I guess their anti cheat can't be like video game DRM but how do you declare someone's moved to be cheating without proving they're using a 3rd party?
Here's how they describe it. https://www.chess.com/article/view/chess-com-fair-play-and-c...
It's a set of statistical methods as far as I can gather, that look at things like timing tells as well as accuracy compared to engine moves and so on. Obviously it's going to be imperfect.
In the same way mining a bitcoin proves that you've computed a large number of SHA-256 hashes, playing certain sequences of chess moves is attestation that you've run a large number of chess calculations—that you've solved a certain large system of constraints that has no real solution other than brute force. Naive cheating (that thing you describe) is obvious, because the cheater is running 6 orders of magnitude more computational power than a human and is proving it to you.
I was going to ask what possible motivation this dude had, but you’ve answered that quite well.
To me that is comically atrocious way to cheat... Like was there any time spend to think more sensible approach?
some basic level of formal culture in chess went missing. Chess is a battle in a war,fought with the mind,and conforming to the ancient rules of engagement and decorum. Certainly in the recent past,the outcome of a match between the representitives of adversarial nations was treated as proof of the intelectual and educational prowess of the winning side. What we are watching is the inevitable consiquence of comodifying everything,and the ho hummyness of the incident bieng disscused is the sound of the door closing.
This guy is 2656 and the world number 69. You might think, meh, so what who's heard of the world number 69 at, say, tennis. But if you are a serious chess player you understand that is an astronomically high level that is completely out of reach for 99.9% of people even if you love the game and devote many years of effort to it. (No doubt the same sentiment would occur to a serious tennis player in the tennis analogy).
I have two points really A) As a chess player and a programmer, I think many a programmer fancies themselves as being a rock star performer but if there was an equivalent of Elo ratings for programmers it would likely be a very sobering experience for almost everybody [I think I am a highly skilled programmer, but my chess rating is 1867].
And B) It is consequently a real personal tragedy for this guy that he has foolishly flushed his career and reputation at age 22. What a waste.
For reference, chess skill has a function indicating likelihood to beat a player of another rank (I don't recall the ratios). This user's chess skill dwarfs mine (1400s), might be higher than anyone I've ever played with in person. I can stomp all my acquaintances easily, but this person might never lose a game to me, no matter how many we play. There's something deeply gratifying about the formalization of chess skill like this imo, it keeps everyone humble. It's good to meet other players, cheers.
That said, I'm only surprised that such a high-ranking player was caught so easily with cheating. I've been expecting for many years that the problem of catching cheating in chess is complicated by motivated cheaters being necessarily intellectual, therefore hiding it well. We may never know our best cheaters, at least during their professional careers, and it's a little disheartening knowing even top100 players can overtly toss away their prospects this way. It would be a surprise for anyone this high to not be invested in chess in multiple financial avenues, so this really is like shooting yourself in the foot.
I remember the Magnus/Niemann controversy from 2023 - that was quite a drama... https://en.wikipedia.org/wiki/Carlsen%E2%80%93Niemann_contro...
This is almost comical. Why did he want to be caught so badly? Or did he stack all of his skill points in chess, to the detriment of common sense?
Unfortunately being good at chess doesn't make you smart. It just makes you good at chess.
There’s a reason D&D has intelligence and wisdom as separate attributes.
if you are a chess grand master then you are also smart. leaving evidence behind can be attributed to nervousness or being a bit scatter brained.
If you're a chess grandmaster then you're good at a game.
* Mr. Kramnik has entered the room *
I still want to see Magnus Carlsons neural net applied to another domain. IF it supports generics