Chat GPT failing on math reasoning task

[josephfrusetta]

We’ve recently had some great topics that discuss how and why ChatGPT/GPT-3 fail at various types of math problems:

We have a problem with basic maths

Hey everyone. As brilliant as GPT-3 seems to be with everything, and how proud people seem to be that it “taught maths all by itself”, it actually has a problem with performing basic math operations. [Screen Shot 2022-12-05 at 13.13.55] [Screen Shot 2022-12-05 at 13.14.20]

Is GPT-3 good at math? Let the answers speak for themselves!

Codex can solve some simple non-linear control problems. It is a matter of how you script it I guess. Look at Wolfram Alpha, they use an AI model too. Reality is stochastic, thus we can solve anything with math.

The short answer is that ChatGPT and GPT-3 are language models. They are meant to predict the next series of words and cannot complete certain types of math problems without additional resources.