I've been using ChatGPT 4.0 for my writing assignments, and I'm hitting a wall with a specific issue: no matter how many times I ask it not to use em dashes (—), it just keeps adding them back in. I always start our chats by stating my preferences clearly, even copying and pasting the em dash symbol and telling it to avoid using it. Initially, it complies and replaces them with commas or semicolons. However, as we move on to different paragraphs or tasks, the em dashes reappear like nothing happened. I've tried being super clear, emphasizing that using em dashes could seriously hurt my grades, but it always apologizes and then continues to use them. I'm baffled as to why this is such a consistent problem. Has anyone else dealt with this or found a workaround?
1 Answer
It seems like em dashes are just baked into how the AI is trained. Maybe it's a design choice to ensure certain writing styles, or it could be tied to the training data. Regardless, it's frustrating when it contradicts your requests! Have you thought about using alternative wording instead of directly stating 'no em dashes'? Could be worth a shot!
Related Questions
xAI Grok Token Calculator
DeepSeek Token Calculator
Google Gemini Token Calculator
Meta LLaMA Token Calculator
OpenAI Token Calculator