You might’ve noticed that ChatGPT — and AI in general — isn’t good at math. There’s a reason, and it has to do with how modern AI is built.
Basically, they’re autocorrect on steroids. Which some of us have been saying for, like, ages.
There’s zero ability to compute problems. Only statistical word salad. AI is a scam.
LLMs (I refuse to call them AI, as there’s no intelligence to be found) are simply random word sequence generators based on a trained probability model. Of course they’re going to suck at math, because they’re not actually calculating anything, they’re just dumping what their algorithm “thinks” is the most likely response to user input.
“The ability to speak does not make you intelligent” - Qui-Gon Jin
Why can’t this person working on his linguist degree not able to do high level math? It’s not their specialty.
To be fair, even someone with a linguist degree knows basic math. GPT can’t even get that right. That’s the biggest problem (and red flag).
GPT can’t get basic math right? That hasn’t been my experience whatsoever.
It can absolutely do basic math. Give us a ‘basic’ math question that it can’t solve.
I kind of want to see something on this because my past experiences with ChatGPT is that it was horrible at math, but I used it again the other week for some insight on some math, and it worked it out completely correctly. I don’t know anymore, is it shit at math, is it good at math or in between?
Its generally pretty good. It is able to walk through and explain its steps very well. The main problem with math is sometimes it will just do something that makes no sense. You basically have to redo it by hand to make sure it got the correct answer. But it gets it right significantly more often than not for anything Ive thrown at it recently, and so its still useful even it requires you to check its work. In the past, though, it was a complete waste of time to ask it to any math.
deleted by creator
Programming languages are structured and have rigid syntax that fits well in a LLM model, so it spitting out working code for simple things is like having a sentence that is structured like a normal person.
The code might not do what you are actually trying to do, or might work while being inefficient, even if it runs.
Both of those are user issues.
Ive heard the chatgpt math problem was fixed in the new one by having it write a python code to complete the math problem and then providing the answer when the code is run.
