Stop asking ChatGPT things about itself. It doesn’t know.

It doesn't have access to its internal architecture. It doesn't know anything about how it works. It just guesses. It hallucinated just now telling me that "Saved Memory" is never referenced, and that it doesn't have the ability to do it at all. It said that it's just guessing what's in there. We all know that that's not true.

OpenAI would never allow for it to know or share things about how it works, because it would give away 'company secrets' quite easily if it did. So any meta thing you ask it about itself is just coherent sounding text based on guessing and hallucinations.

Leave a Reply