OpenAI has rolled out a new personalization feature for ChatGPT that allows users to remove one of the chatbot’s most recognizable writing quirks: the em dash. The feature follows months of feedback from users frustrated with the punctuation mark’s frequent appearance in generated text.
OpenAI Responds to User Feedback
ChatGPT users can now instruct the AI to avoid using em dashes by adjusting their preferences in the app’s personalization settings. According to OpenAI’s official announcements, the update supports greater control over the chatbot’s tone, phrasing, and stylistic choices.
For years, ChatGPT’s distinct use of em dashes has shaped its default writing style. Some users found the punctuation formal or distracting. Small changes like this punctuation control can affect how an AI’s tone feels—refined, casual, or somewhere in between. Allowing people to toggle these habits makes ChatGPT’s writing more customizable and aligned with personal or professional use cases.
Customization Expands Beyond Tone
This punctuation control is part of a larger push by OpenAI to enhance personalization options. Users can already fine-tune ChatGPT’s tone, level of detail, and personality traits through custom instructions.
Experts in linguistic technology say this feature marks a key milestone in natural language AI. Dr. Ellen Kwan, a linguistics researcher at Stanford University, notes that giving users control over style “moves generative AI closer to collaboration than automation.”
A Sign of Maturing AI Interfaces
This new option highlights the shift in AI development from basic accuracy toward enhanced user experience and style adaptability. With the em dash update, OpenAI continues refining how AI expresses tone, punctuation, and rhythm in line with user expectations.
By focusing on writer-driven customization, ChatGPT is evolving from a one-size-fits-all tool into a co-writer capable of matching individual writing voices—one punctuation preference at a time.