0
I’ve tried asking ChatGPT the same question more than once and received noticeably different wording, examples, and sometimes different levels of detail. This can happen even when I use the same model and don’t intentionally change the prompt.
What parts of the response process can vary between attempts? I’m especially interested in whether starting the same prompt in a new chat makes a meaningful difference, and how much settings or conversation context affect the result.
