What happened? One of the biggest problems with ChatGPT has now been fixed. Sam Altman announced in an X post that ChatGPT will now comply when specifically telling users not to use dashes in the Custom Instructions tab. By adding a rule to the custom instructions to avoid using dashes, you can finally get ChatGPT to stop using them.
- The update addresses a long-standing complaint that ChatGPT’s heavy reliance on Em-Dash caused its output to look “bot-written.”
- The change only works if the setting is set in Personalization or Custom Instructions settings. Saying this verbally in a request may not ensure compliance.
- Altman said this is part of a broader push toward customization: ChatGPT already supports personality presets and save features, and this is one of the earliest “style” controls introduced.
Small but nice victory:
When you tell ChatGPT not to use dashes in your custom statements, it finally does what it’s supposed to do!
— Sam Altman (@sama) November 14, 2025
Why this is important: Punctuation may seem negligible, but it plays a crucial role in the tone, readability, and professionalism of the writing. For many users, such as writers, students and professionals, the dash became a flashing signal that a text was AI-generated, undermining confidence in ChatGPT’s usefulness for “human-style” writing. This update shows that OpenAI understands that users don’t just want accurate content; They want style control. The ability to customize the way ChatGPT writes (tone of voice, slang, punctuation) makes it more like a real writing assistant than a generic text generator.
However, the solution also raises larger questions: If the model has struggled with a punctuation mark for years, what other stylistic quirks remain? And how many of those require user-specified rules rather than an inherent AI improvement? This could be a step toward deeper AI personalization, but also shows how behavioral quirks persist in large-language models.
Nadeem Sarwar / Digital Trends
Why should I care? If you use ChatGPT to write important things like emails, reports, scripts, essays, or social media posts, this update gives you more control over how the AI sounds, not just what it says. Style consistency is important, and removing obvious “AI tells” like excessive use of dashes can make output feel more natural and personal. It’s a small change, but it signals a larger shift in allowing users to shape their AI’s personality.
- It reduces the amount of editing you have to do by eliminating one of ChatGPT’s most well-known typos.
- This makes the AI behave more like a true, customizable writing assistant rather than a unit generator.
- It hints at more granular personalization tools in the future, giving you greater influence over tone, formatting and style habits.
Okay, what’s next? Now that ChatGPT can give up its em-dash habit on command, the real question is how far this new obedience will go. If OpenAI continues to push in this direction, we may soon see deeper style controls: from sentence length to emoji tolerance to whether the AI should sound like a calm assistant or someone who has had three too many coffees. This update also brings personalization back into focus.
Therefore, you can expect creatives, students and companies to test how configurable the model really is and whether these little switches can lead to an AI that finally blends in with your texts instead of standing out. It’s still early days, but if ChatGPT continues to take notes so well, your future drafts may look a lot more “you” and a lot less “AI.”
Comments are closed.