Re:Builder.ai did not "fake AI with 700 engineers"
by ben_w on Friday June 13, 2025 @11:17AM
(#44269321)
> As the recently released Apple study demonstrates
The Apple study that did Towers of Hanoi and concluded that giving up when the answers would have been too long to fit in the output window was a sign of "not reasoning"?
The Apple study that did Towers of Hanoi and concluded that giving up when the answers would have been too long to fit in the output window was a sign of "not reasoning"?
https://xcancel.com/scaling01/status/1931783050511126954
I mean, on that basis, anyone who ever went "TL;DR" is also demonstrating that humans don't reason.
> That's not "intelligence" however you define it because they can only solve things that are already within their training set.
This is proven untrue by, amongst other things, looking at them playing chess. They can and do play moves not found in the training data: https://www.lesswrong.com/posts/yzGDwpRBx6TEcdeA5/a-chess-gp...