I guess that's way more complex because there is no one right answer. (Interesting use case btw) And if you ask gpt for an specific answer to something it can'

I guess that's way more complex because there is no one right answer. (Interesting use case btw) And if you ask gpt for an specific answer to something it can't actually know for sure it gets weird fast. Asking it to guess how much of what it told is based on actual things it 'knows' at least seems like a reasonable question for a LLM with only one right answer.