HenryBemis
10 months ago
I was wondering (also to billconan's comment), does anyone check that the LLM doesn't hallucinate? I mean, if one reads a study for the fun of it, go crazy! If someone finds a study about diabetes, and decides to follow the 'diet' that the book suggests, it can go very south - very fast.
melvinmelih
10 months ago
This is a valid concern and while there are no guarantees that the AI won't hallucinate (hence the disclaimer in the book, especially for medical topics), I try to minimize it by pairing the writing with real-time research from Perplexity, so at least it is (or should be) based on verifiable information.