I'm struggling to get ChatGPT to stop using the phrase "it's not X, it's X" in its responses, no matter what I try. I even tried banning the word "not" altogether, but it seems to ignore that too. For example, when I ask it to analyze its previous answers based on memory, it realizes there's an issue and suggests starting over instead. I'm just curious why it's so fixated on this language pattern!
4 Answers
It’s kind of like the pink elephant effect—once you mention something, it seems to pop up more! Try rephrasing your prompts or avoid mentioning the phrase directly; that might help steer it away from using it repeatedly.
I've seen a lot of posts lately that seem full of ChatGPT’s quirks, especially on LinkedIn. It's interesting how popular they still are, even if they aren't the best quality.
It's definitely frustrating! I find it super annoying too when it keeps repeating that structure. You're not alone in this; even I've noticed it happening everywhere!
Have you tried giving it very specific instructions? I mentioned to avoid "rhetorical symmetry" and it seemed to change how it responded for me. Maybe something similar could help!
Related Questions
xAI Grok Token Calculator
DeepSeek Token Calculator
Google Gemini Token Calculator
Meta LLaMA Token Calculator
OpenAI Token Calculator