Chess will never be solved, here's why

just down to accuracy of the question. I asked a more specific question:

"Is it realistic that computing would advance enough to fullfil the requirements for a weak solution of chess?"

Which makes it consider what can be realistically possible in terms of technological advancement. Also I specified 'weak solution' as opposed to strong. If you form the question like "can chess be solved" it might consider the question by todays tech standards.

But I agree LLM isnt the way to answer these questions, I dont enjoy AI enhanced debate or especially just dumping AI responses in a thread.

No, I don't believe we should.

The definition was fine, IMO. Slightly loosely worded, but consistent with understanding.

Compare and contrast.

A game is considered "weakly solved" when, for any given position, an algorithm can determine a strategy that guarantees either a win or a draw for one of the players, assuming perfect play from that point onward. - Grok 3

Weak solution
Provide an algorithm that secures a win for one player, or a draw for either, against any possible play by the opponent, from the beginning of the game. - Wikipaedia

The second definition does not require an algorithm for any position for a weak solution of a game. There are alternative incompatible definitions of "weak solution of a game" in game theory literature, but none that I've seen that do require such a strategy for any position. 

On the other hand the Wikipaedia definition of strong solution (along with others) does require such strategies for any position (at least for any legal position if you feel there is a distinction).

Strong solution
Provide an algorithm that can produce perfect play for both players from any position, even if imperfect play has already occurred on one or both sides. 

The Grok 3 definition is simply wrong with any generally accepted meaning of the term.

@Martin

yes that particular sentence by Grok 3 was kind of incomplete but it follows it up with this, specifying optimal strategy "from starting position" and not all possible positions in chess

"For chess, this would mean finding a strategy that, from the starting position, ensures at least a draw (or possibly a win) for White, and similarly for Black in response."

Yes. There it disagrees with itself in the same paragraph.

@Dubrovnik-1950 didn't ask, "can chess be solved?". he asked, "will chess ever be solved?". His question shouldn't be considered by today's tech standards.

In just the same way as if you were asked fifty years ago, "can a chess program beat a grandmaster" or, "will a chess program ever beat a grandmaster", it would have been invalid to base your answer to the second question on the tech standards of the time.

@Martin

I know for us its clear but with these chatbots sometimes you have to be extra specific to make it consider all options. But I have no doubt the answer can vary even with the same exact question, especially when its speculative in nature...

You were a little disingenuous by omitted the last part of the Grok 3 definition that was very similar to yours! But I agree the wording was substandard (it would be bad for anyone who doesn't already know the meaning).

The full text was this (my emphasis).

<<A game is considered "weakly solved" when, for any given position, an algorithm can determine a strategy that guarantees either a win or a draw for one of the players, assuming perfect play from that point onward. For chess, this would mean finding a strategy that, from the starting position, ensures at least a draw (or possibly a win) for White, and similarly for Black in response.>>

I can see how the first part can be interpreted as a strong solution if it means it applies to ALL given positions, but it can also be interpreted as a weak solution of a game defined by a given starting position and a given set of rules for moving and ending. This is the interpretation carried through to the last sentence.

But while one can interpret the first sentence in light of the second, it is badly worded. Better is something like:

"A game with a given starting position is considered "weakly solved" when, an algorithm can determine a strategy that guarantees either a win or a draw for one of the players, assuming perfect play from that point onward, and the other player has a strategy guaranteeing the same result".

The omission of the latter is important. If the first player has a winning strategy, exhibiting a drawing strategy for one player (and a losing one for the other) is not a weak solution. Indeed the "similarly for black in response" seems to preclude any winning game, since "similarly" does not include the case when a player is losing!

But better to just let them chat with each other somewhere else and leave room for sensible discussion.

I was not being disingenuous in any way. Grok 3's definition of "weakly solved" (purportedly for any game) is the sentence I quoted, contradicting @OctopusOnSteroids's assertion that it understood the term. The fact that it then goes on to misapply it's own definition to chess, doesn't affect that.

As for your interpretation, that's precisely why I highlighted the term "game". There can only be one interpretation of "any position in a game". It doesn't talk about a weak solution of a position anywhere (and I haven't seen such a definition in the literature, though I've used the term myself in past posts).

I agree with you Martin but please dont direct that my way (maybe you didnt), as I merely posted the Grok response to demonstrate that the bot that received much praise here does not necessarily agree with those praising its incredible wisdom. Thats about it.

It made a mistake in that particular sentence and youre welcome to point that out, but it understood the term in the context the discussion was being had (weakly solving chess), which is sufficient for my purpose to prove its response was relevant to weakly solving chess. Thank you.

I agree its statement about chess is unexceptionable (except for the clause, "For chess, this would mean ...", which is wrong - it wouldn't).

No, I didn't.

@Elroch re #19251. @OctopusOnSteroids

Well actually, now I read it, the second sentence is b*llocks as well. The can't both have at least a draw if the position is a win.

I'm bloody working on it. I know what I'm doing. You know sweet fa. Shut up and let me get on with it. I don't post two second output regurgitated from Grok3.

As I already said, patience! Have you got any? The less you pester, the sooner you'll get it.

Because of "similarly for black in response"? Come on now we are nitpicking arent we

Imagine not understanding basic probability theory while trying to argue about the architecture of chess