Why I'm still bearish on LLMs after Navier-Stokes • Hacker News Board

Re: Why I'm still bearish on LLMs after Navier-Stokes

#661

Post by lostmsu » Thu, Sep 17, 2026, 2:44 PM UTC

HarHarVeryFunny wrote: post #602

lostmsu wrote: post #594

Earlier quoted context omitted.

> prediction which is closer to memorization > don't memorize inputs - they predict them I feel some tension here. > rice grains on a chess board? Sure, but this has nothing to do with chess, and nothing to do with how many games were in the LLM's training data. > just a list of 64 numbers > remember even a few positions? Sure they could, but that's irrelevant. I don't think you do. Or rather you do know the legend b…

You are talking about 2^64 being a huge number I assume ? If not, then what are you talking about ? If yes, then what is the relevance to an LLM playing chess ?

> just a list of 64 numbers

> remember even a few positions? Sure they could

A rough estimate of number of positions across all X move games is X^10. For 15 moves it is hopeless to remember even a relatively small part of them. Typical game has 40 turns, 1 move per player, so 80 moves.

Re: Why I'm still bearish on LLMs after Navier-Stokes

#662

Post by freejazz » Thu, Sep 17, 2026, 3:23 PM UTC

0c3ca83 wrote: post #499

reedlaw wrote: post #494

Earlier quoted context omitted.

Because agents lack human judgment. At the very least there's a need for a human-in-the-loop with agentic processes. Otherwise, it's like running a coding harness with --dangerously-skip-permissions all the time.

Why do you think judgement is impossible to automate? What aspects of it do you think make it hard?

I think right here you're demonstrating that judgment doesn't come easily

Re: Why I'm still bearish on LLMs after Navier-Stokes

#663

Post by HarHarVeryFunny » Thu, Sep 17, 2026, 3:41 PM UTC

lostmsu wrote: post #661

HarHarVeryFunny wrote: post #602

Earlier quoted context omitted.

You are talking about 2^64 being a huge number I assume ? If not, then what are you talking about ? If yes, then what is the relevance to an LLM playing chess ?

> just a list of 64 numbers > remember even a few positions? Sure they could A rough estimate of number of positions across all X move games is X^10. For 15 moves it is hopeless to remember even a relatively small part of them. Typical game has 40 turns, 1 move per player, so 80 moves.

1) The number of unique chess games that could theoretically be played (but mostly never have been), is irrelevant to what an LLM is remembering. It can only remember what was in it's training data - a far smaller number of maybe 10's of millions of games (of 30-50 moves each).

2) An LLM is not going to memorize vs generalize when there is no training pressure to do so. You might expect it to memorize book openings that occur over and over in the training data, but not some random non-celebrity game that occurs once in the Lichess dataset and is never again referred to.

> They can't possibly remember even a few positions. Don't you know the legend about rice grains on a chess board?

If the wise man was a bit wiser, he'd have asked for his rice on a snakes & ladders board (100 squares, not 64) and would have had 2^36 more rice, which is equally irrelevant.

Re: Why I'm still bearish on LLMs after Navier-Stokes

#664

Post by lolakutty » Thu, Sep 17, 2026, 4:59 PM UTC

moomoo11 wrote: post #653

lolakutty wrote: post #641

Earlier quoted context omitted.

Your use case totally makes sense. The non-sense is the part where you think putting agents in a loop is going to yield better results indefinitely...

i guess me and my customers are morons then and you’re a genius when you’ve not achieved anything life of a w2

I meant no such thing about your customers.