Hacker News Reader

Comment by ttoinou | Hacker News Reader

That is a classic riddle and could easily be part of the training data. Maybe if you change the wording of the logic, then use different names, and change language to a less trained on language than english, it would be meaningful to see if it found the answer using logic rather than pattern recognition

View on HN

Had you paid more attention, you would have realised it's not the classic riddle, but an already tweaked version that makes it impossible to solve, hence why it is interesting.

Mellowobserver above offers three valid answers, unless your puzzle also clarified that he wants to get all the items/animals across to the other side alive/intact.

Indeed, but, no LLM has ever realized that I don't actually specify that he can only take one thing at a time. It's natural that it would assume that (as would most humans) because it would be so heavily primed to fill that in from every other version of the puzzle it's seen.

I'd give them full credit if they noticed that, but I was also wanting to see if, given the unstated assumptions (one thing in the boat, don't let anything eat anything else, etc) they'd realize it was unsolvable.

Why is it unsolvable? I am confused.

The unstated assumptions, common in all the other puzzles of the kind, are that the farmer wants to bring everything across, nothing is to be eaten by anything else, everything must be brought over by boat, and the farmer can only bring one thing in the boat at a time. Anything left with something that will eat another thing will eat it.

This variation of the classic puzzle is unsolvable. If you have a solution, let me know.

most humans would not assume that since most humans are not heavily primed by logic puzzles.

All those answers recognize that its a trick though!

Both Claude 4 Sonnet and Opus fail this one, even with extended thinking enabled, and even with a follow-up request to double-check their answers:

“What is heavier, 20 pounds of lead or 20 feathers?”

Can humans answer this correctly ? It is ambiguous

chatgpt (whatever fast model they use) passed that after i told it to "read my question again"

Ah right. But maybe someone thought about this simple trick / change already too.

Reply on news.ycombinator.com