The Wall Street Journal reports.
The report says that Chess.com uses a variety of cheat-detection tools, including: analytics that compare moves to those recommended by chess engines; studies of a player’s past performance and strength profile; monitoring behavior such as players opening up other browsers while playing; and input from grandmaster fair play analysts.
Computers have “nearly infallible tactical calculation,” the report says, and are capable of beating even the best human every single time. The report says dozens of grandmasters have been caught cheating on the website, including four of the top-100 players in the world who confessed.
Identifying violations in over-the-board games remains a major challenge. The main reason is that grandmasters who cheat require very little assistance. For a player operating in elite circles, a couple of subtle moves in critical spots can be enough to tilt the balance against a world champion. That makes definitively proving allegations of cheating difficult unless a player is caught in the act—by using a phone in the bathroom, wearing a small earpiece or receiving signals from someone in the audience....
Posts
34 comments:
Chess became obsolete for non-face-to-face human competition in 1997 when IBM's Deep Blue beat the world champion:
https://www.ibm.com/ibm/history/ibm100/us/en/icons/deepblue/
It's no surprise that people cheat with personal computers that are 25 years more advanced. Believe Elon Musk when he speaks about the dangers of artificial intelligence.
Computers killed this kind of chess. I had a book of chess problems by a grandmaster, ran several of them though a chess engine, Stockfish, I downloaded for free, and the chess engine demolished the grandmaster's analysis of any of the problems he rated as difficult.
Not to mention that there is a massive amount of memorization and it may be a lot of moves before the players get to "new chess." Chess is one of those things where the charm wears off, the more you know.
In fact, this grandmaster took problems from his own matches, sometimes, and one of them was a position whee the other guy resigned, but if you put the setup into Stockfish and played the suggested moves for both sides, the other side won. Google's AlphaZero could possibly beat the open source Stockfish. That's where the big competitions are going. Battles of AI vs Intentionally programmed software like Stockfish, or AI vs AI.
Why is it that every time a chess grand masters plays a game with a dido up his butt people assume he's cheating? Maybe he's just lonely.
Tyler Cowen is a great source for updates on all this.
I'd love to hear what Kasparov thinks. Almost everybody remembers the Deep Blue exercise and to me, it wasn't all that long ago. A lot of the near future is going to look like this as the AI/ML/brain interface develops and matures. The exercise may push humanity over the edge, but if it doesn't it may mean that Kurzweil's musings are not as crazy as they sound.
1961 - "Only her hairdresser knows for sure!"
2022 - BZZZ BZZ BZ BZ BZZZZZ - Q to c5.
To what exactly is this analogous in the development of the species on this planet?
Printing? Speech?
i don't understand?
WHY Couldn't he 'self identify' as a computer? Wasn't he mis assigned at birth?
Isn't his use of a computer, JUST LIKE hormone replacement therapy?
Wait a MINUTE! i just realized, that Hans Moke Niemann is a disgusting white cis male!
Disgusting white cis males are the Most Reprehensible things ON EARTH;
and there can be NO Justification for the disgusting CRIMES he has committed.
Maybe the kid just likes vibrating anal beads. That's a thing now, isn't it?
It's not AI. It's just a chess program. The cleverer the program, the better it does.
Imagine a catalog of every possible position and the best move for each. A chess program just does a more efficient job of coding that.
And recently there was cheating at poker when a full-breasted gal improbably called an all in with just jack high. What’s next? Cheating at golf?
- Krumhorn
A lot of the near future is going to look like this as the AI/ML/brain interface develops and matures. The exercise may push humanity over the edge...
ha ha ha
What pushes me over the edge is when fucking Blogger can't do the fucking task of publishing posts once and then moving on.
Push the button, Blogger! You can do it!
This grips me more than would a muddy old river or reclining Buddha and thank God I’m only watching the game, controlling it.
Have you ever gotten into an argument with Siri? That is annoying as shit. Siri doesn't have to turn into HAL to be a pain in the ass. Just her normal robot brain does the trick.
My family was lost in South Carolina, we had no idea where the hell we were, looking for signs, anything. And fucking Siri wanted us to turn right, off the highway, into a dirt road leading into a cornfield. I am not kidding. No help at all. Making it worse.
So consider the possibility that A.I. never goes insane, it's just so fucking logical and inhuman that we go insane dealing with her. The key, the absolute key, as far as I'm concerned, is never giving A.I. any power.
Would you trust Blogger with the fucking nuclear launch codes? I trust my dog more than Blogger. No help at chess, of course, but she's way nicer and closer to humanity.
That's not why Bobby Fischer went paranoid, but it would be a good reason for chess players to go off the deep end.
Soros, Schwab, and now Hans Moke Niemann.
Bond villains are real all of a sudden.
"...including: analytics that compare...Computers have “nearly infallible tactical calculation"
The same efforts were put to use on the 2020 presidential vote, and despite clear evidence of fraud, "no evidence of fraud."
For me, the big tell was Magnus Carlsen (Carlson?), the reigning world chess champion, who has been chess champion for something like a decade, when he said that Neimann was far too relaxed after their game. Said that he plays plenty of chess prodigies, but they all focus and are concentrated and clearly working hard (as they should! When you go up against the GOAT to try to win, you are trying to win! And Magnus has a pretty good claim to be the GOAT of chess; or at least top 5 all time). Neimann, however, was far, far too relaxed--almost bored -- when they played and Neimann beat him with ease. Carlsen, like all other humans, cannot beat modern chess software when it's at full strength.
This has been quite the juicy scandal. Chess drama? Who knew, right? Certainly, the drama speaks for itself.
Can they not use RF detectors to prevent this?
"And fucking Siri wanted us to turn right, off the highway, into a dirt road leading into a cornfield."
I got this from my GPS in Nova Scotia. Its internal map doesn't have speed limits for the dirt roads, so it assumes you can travel the local legal maximum on them, so it always decides they're the fastest way there.
In Puerto Rico, the map was just flat out wrong in spots, resulting in some very long detours.
Mixed doubles is coming to chess.
Computers and their human partners.
Most of the "human interest stories" will have to be about the computers.
'...monitoring behavior such as players opening up other browsers while playing...'
I read this article already this morning.
Who would open another browser to cheat?
Wouldn't a smart cheater have 2 computers?
'Maybe the kid just likes vibrating anal beads. That's a thing now, isn't it?'
I saw Vibrating Anal Beads open for Prince at the Fillmore in '98.
They were good, but the lead singer seemed stuck up...
"Imagine a catalog of every possible position and the best move for each. A chess program just does a more efficient job of coding that."
I don't think that's true, rh. If it were, then chess would be solved. There would be a known best first move for white, and a known best response to any move black can make, and a known longest possible game. And that would be that.
People were attacking Carlson but he ended up being right. These guys, Grandmasters playing at the highest level, can sometimes tell is someone is cheating because of the time spent on complicated situations. Computers used to lose to Grandmasters, because the GMs instantly realize their are only 1-4 possible correct moves in a situation, while computer has grind through all the variations.
So, when a GM looks at a complicated situation where there are 3 good moves, and takes an absurdly small amount of time to select the best one, they suspect cheating. From what I can understand that's what happened with carlsen and the Nieman. He made a fantastic move and only took 17 seconds to select it.
People went back and looked at his games and showed he was making the best possible move per the computer 90 percent of the time. The same analysis showed that Carlson or Fisher only do it on average 72 percent.
Chess is actually better than ever. online chess means you can play 24/7 against people at your own level. GM's are better than ever, and they've gotten around the tendency for draws in classical chess by adopting blitz games. They'll probably stop the cheating by time delay and not allowing spectators.
Fluoroscopy the players just before they sit down to play. Simple enough to do.
My theory: the more we are "surprised" that there's is cheating going on, the easier it might be to cheat. Whereas, if cheating is expected you might recruit "stakeholders" to aid catch it.
Casa Blanca clip
"Computers have “nearly infallible tactical calculation,” the report says, and are capable of beating even the best human every single time."
So basically, Chess is now nothing more than a sad bit of nostalgia. If computers win all the time, then the best human is the most computer-like, i.e., pattern recognizer. Chess Grandmasters going the way of "Mr. Memory" in the face of computers.
Reminds me of the story about Napoleon and the chess-playing Mechanical Turk. Supposedly the emperor attempted an illegal move and the Turk refused to continue.
How do programs/AI do with 'kriegspiel' chess?
Incomplete knowledge and ambiguity--a fog on the board . . .
US Grandmaster cheats -
Media reports blaming "Russian disinformation" in 3... 2... 1....
Chess is a big deal here in Norway because of Magnus Carlsen. His matches are shown on national TV stations and every time he's been in the World Championship it's a major event.
So when he accused Niemann of cheating when he lost to him playing white it got a lot of coverage. Some thought it was wrong to make such an accusation without solid proof. Kasparov was critical. But Magnus was correct. He's really in a league of his own when it comes to chess. And he's also a very well adjusted and sensible person. Unlike Bobby Fischer, he had a family that understood the difficulties that having a mind like his causes. There's a good biography of him. Here's some excerpts:
https://www.youtube.com/watch?v=TW0yJYab5YA
I'm a bit surprised that 3-dimensional chess never really caught on. My brother and I were briefly into chess (as much as we were briefly into backgammon) and we got a 3D chess set for Xmas one year and played it for a couple months before moving on to something else.
It's a very strange game that totally breaks your head.
AI is not intelligence, it’s just a program that the programmer no longer understands, that’s why I contrasted it with “intentionally programmed.” AI is the term we have.
I played a lot of on-line chess in the period from about 2002-2014. Cheating became more and more prevalent over that time. It is just something you have to accept because there is no way to stop it. It never really bothered me because I wasn't playing for money and I didn't give a shit about rating points- I would often play an online chess engine just to work on my game since it is one way to play the very best opponent available.
Now, at the elite level, there is something you can do about it- you ban someone from money on-line tournaments who is a known cheater And for over-the-board tourneys, the players should be scanned for metal devices and bathroom breaks get monitored scrictly. I am unconvinced that Niemann cheated in the game against Carlsen, but I understand why he isn't getting the benefit of the doubt given his history of cheating online.
I’m curious who would win if one super computer squared off against another. My now deceased friend, grandmaster Emory Tate (father of the newly-famous Andrew Tate who I’ve had in my home), told me the theory was that once chess was “solved” black would win. He said to move second ultimately is an advantage because once the first move is made, predestination takes over.
I’ve often wondered if he was right. Emory died, btw, doing what he loved: he was in the middle of a chess game.
“I’m curious who would win if one super computer squared off against another.”
Google Stockfish v. Alphazero—lots of games online. It’s interesting to see engine analyses of “obsolete” openings like the Evans Gambit—new lines are found that can be used by humans now that computers have shown the way.
Post a Comment