What kind of question is that? I wrote something from memory, then as an afterthought fed it through a system that is quite good in giving cursory, fact-based review if its validity. You see the whole history and know the system prompt isn't something malicious against the context.

I don't like ChatGPT's biases in many things either but being that hard against it while it cites Reuters etc isn't really sensical.