OpenAI introduced a new writing-style feature for ChatGPT Work that can study how users write and apply those patterns to future responses. The feature is designed to make AI-generated text feel more personal and closer to the user’s natural voice.
ChatGPT Work can identify details such as frequently used phrases, email sign-offs and unusual capitalization habits. It can learn these patterns from connected work services, including Gmail, Google Drive, Slack and SharePoint.
The feature is aimed at people who regularly use ChatGPT for workplace writing. Users can connect the tools they already use and allow ChatGPT Work to reference their emails, messages and files.
OpenAI described the feature in an X post, saying, “Connect the tools you use every day, like Gmail, Google Drive, Slack, and SharePoint, and ChatGPT Work will learn your writing style from your emails, messages, and files and carry it into whatever you write next.”
The system can pick up small habits that may otherwise be difficult to describe in a prompt. These include preferred expressions, the way a user closes messages and their approach to capital letters.
This could reduce the need to repeatedly explain a preferred tone when asking ChatGPT to prepare an email, document or message. The learned style can instead be carried into later writing tasks.
Users can set up the feature through Settings > Personalization > Writing style in ChatGPT. They can then connect the work tools they want ChatGPT Work to use for understanding their writing patterns.
Once activated, the writing style applies to messages across ChatGPT Work on the web and mobile apps.
The feature is available on the web for paid plans that have access to ChatGPT Work. Its focus is not simply on changing the tone of an AI response. It aims to make generated content reflect the specific habits that make each user’s writing recognisable.
The move gives ChatGPT a more direct way to personalize workplace writing. It also signals a broader shift towards AI tools that adapt to individual communication habits instead of relying only on generic writing preferences.
Also Read: GPT-6 Astra Launch Goes Wrong: Sam Altman Says Sorry to ChatGPT Users