ChatGPT in a nutshell
Well, sort of
It does generate nonsense, but unlike Calvin the ChatGPT is generating nonsense based on nonsense sample data, so Calvin’s is still better.
I know hating ChatGPT is trendy, but while I think this AI boom is absolutely idiotic and LLMs aren’t suitable for a lot of the things people try to use them for, I think there’s a real tendency for people to make it seem like everything about them is garbage. Pretending that even their training data is “nonsense” is just silly
It’s not “trendy,” if anything liking it is “trendy,” hating it is the educated stance.
Clearly you failed to understand the “prompt” because the context in which we’re discussing this is supposed to be about intentionally creating nonsense.
Unfortunately the average person prefers flowery language for some reason. So that’s what OpenAI optimized for.
If you tell it to be precise and short it usually works fine.
