On October 7, OpenAI announced Intelligent UI in ChatGPT, allowing GPT-6 to choose how an answer works as well as what it says. The rollout started on October 7 for Plus, Pro, Business and Enterprise. Free and Go are scheduled to start receiving it on October 8. Availability is gradual, so accounts may not receive the feature at the same time.
What changes in an answer
According to OpenAI’s help page, conversations can include diagrams, buttons, forms and other interactive elements. A response can still be plain text when that is sufficient. The feature works in Chat; it does not extend to Work or Voice.
One practical experiment is to ask for a comparison of two work schedules with adjustable task durations. This gives you a way to judge whether interaction makes the decision easier. Extra buttons may add little to a one-sentence translation: choose the format around the task.
A Pro subscription is different from Pro reasoning
OpenAI distinguishes the ChatGPT Pro plan from the Pro reasoning level. Intelligent UI works from Instant through Extra High, but not with Pro reasoning on GPT-6 Astra. A Pro subscriber can therefore use the interface with another supported reasoning level.
The model documentation also says there is no separate Intelligent UI quota; existing model and tool limits still apply. If an older desktop app does not display the new format, try the web version.
Choosing a simpler layout
On the web, Personalization → Layout and Visuals → Simple reduces visual elements. It does not guarantee text-only answers.
Our practical takeaway is to check whether the response makes its data and assumptions easier to inspect, rather than judging the appearance alone. An interactive budget based on the wrong starting amount is still a wrong budget. For clearer task instructions, see our guide to writing AI prompts.
Join the conversation
Stay on topic and respect other readers. Your first comment may appear after editorial review.