- OpenAI has added a new feature to ChatGPT that lets users interrupt the AI middleman
- Users can type context or corrections to a prompt without restarting by using the “Refresh” button.
- The feature is especially aimed at those using GPT-5 Pro or Deep Research
Accidentally submitting a prompt to ChatGPT only to realize you forgot to include some details or copied the wrong phrase from a website is a common occurrence. Still, it’s a real pain if you’re using the GPT-5 Pro model or ChatGPT’s Deep Research feature because they take much longer to answer and you can have a limited number of prompts to send.
That should be less of a problem now that OpenAI allows ChatGPT users to interrupt the chatbot mid-reply. During the minutes the AI compiles an answer, you can add new information and redirect its reasoning without having to start over. It’s a way to control the conversation in real time.
While the AI is considering how to answer your query, you can click “Update” in the sidebar and add any new context or clarifications, then watch the AI immediately adjust what it wants to say.
No reboot required. You can simply make ChatGPT change its mind while it’s still writing a response now.
You can now abort long-running queries and add new context without restarting or losing progress. This is especially useful for refining deep research or GPT-5 Pro queries, as the model will adapt its response to your new requirements. Just hit refresh in the sidebar and… pic.twitter.com/kESrkU9hc95 November 2025
Practically, the ability to edit mid-response can save a lot of time for researchers and analysts using ChatGPT for deep multi-step queries. Imagine someone writing a data analysis report only to realize that a key source was out of date.
Before this update, they had to stop the model, rewrite their query, and run it all again. Now they can just say “replace 2022 with 2024 data” and see the text update accordingly.
It’s a minor miracle for anyone who’s ever sat through a long, overconfident AI explanation and waited for the model to be finished, just so they could correct a wrong assumption at the start. For a system built on predictive text, this marks a shift from simple back-and-forth messaging to a more interactive format.
Real-time AI adjustments
Reducing wasted time due to errors in a prompt or AI’s understanding of it is definitely something ChatGPT users will like. The update also continues ChatGPT’s gradual evolution into a more adaptive service.
Since GPT-4, users have been able to provide follow-up prompts, but the flow has always run alternately. The continuous, overlapping feedback loop is more useful as you no longer wait your turn to speak. AI listens while it writes.
In a way, OpenAI ChatGPT gives a sense of conversational humility. Earlier models were notorious for plowing ahead even when they had misunderstood a question or misunderstood context. Of course, the update doesn’t make the AI self-aware, but it does enable it to direct itself in a way closer to actual interaction with a person.
And of course you can too follow TechRadar on TikTok for news, reviews, video unboxings, and get regular updates from us on WhatsApp also.
The best business laptops for all budgets
