logoalt Hacker News

airstrike • today at 7:32 PM • 1 reply • view on HN

It's funny because LLMs are actually horrible at logic.

Current AI is actually good at search, but I can see how people mistake one for the other in the context of math, for example.


Replies

rurban • today at 8:52 PM

Which is funny, because LLM's are great in writing proofs for logic solvers. They are excellent at logic.