ChatGPT’s new instruction options included the user’s nickname and profession as well as personality traits for the chatbot.
Some of the suggested personality traits for ChatGPT include “Gen Z”, “Chatty”, and “witty”
Photo Credit: Reuters
OpenAI might have added several new options to its Custom Instructions feature for ChatGPT on Thursday. Several users shared screenshots of these new options in custom instructions that allow users to further personalise the responses generated by ChatGPT. These new options include options to add the user's nickname, profession, as well as personality traits. However, it appears these new options were later removed from the chatbot. It is speculated that OpenAI might have enabled this feature before it was ready to be rolled out.
Multiple users on X (formerly known as Twitter) posted about the new feature added to ChatGPT. Essentially, it is an expansion of an existing feature dubbed Custom Instruction. As per the screenshots shared by these users, the AI firm added new customisation fields to the tool, allowing users to have granular control over how ChatGPT generates responses.
Custom instructions are grounding tools that eliminate the need to type the same prompt over and over. For instance, if a user prefers formal and concise responses with an analytical tonality, they can add this information to the custom instructions menu to ensure that whenever the chatbot responds to a query, it adheres to these rules.
New custom instruction options on ChatGPT
Photo Credit: X/Tibor Blaho
So far, these instructions could be added to a single place in the form of a text prompt. But based on the screenshots shared by users on X, it appears that the chatbot added new options that allowed them to better control the responses generated by ChatGPT.
The users can tell the AI chatbot a nickname for themselves and remember their profession. Additionally, users can also add several personality traits to ChatGPT such as “witty”, “opinionated”, and “Gen Z”.
It is unclear how these traits shape the chatbot's responses, as Gadgets 360 staff members were not able to verify the presence of the feature, and it was not visible on the chatbot at the time of publishing this story.
Users on X also noted that the feature suddenly vanished after a while. As noted by TechCrunch, OpenAI might have accidentally released the feature before it was ready. It is not clear when the feature might be officially rolled out.
Catch the latest from the Consumer Electronics Show on Gadgets 360, at our CES 2026 hub.