Buttons and Charts Land in ChatGPT Answers as GPT-6 Rolls Out From October 7
OpenAI has unveiled Intelligent UI, which embeds buttons, charts, and simple tools inside responses. Paid tiers including Plus and Pro get it starting October 7, with Free and Go following on the 8th.
OpenAI on October 7 widened access to GPT-6 in ChatGPT and announced a new response feature called Intelligent UI. GPT-6 had previously been released to paying customers first. Now, instead of only reading an answer, users can interact with buttons, input forms, and charts right inside the conversation.
How the in-answer tools work
Intelligent UI changes the shape of a response to fit the question. When something needs explaining, it pairs text with visuals. When a question calls for something you can click, it adds buttons or input forms. Tools such as calculators or small games can also be built and used inside the chat. For some questions, it still answers in plain text, as before.
The feature works on both web and mobile. Interface elements appear one after another while the response is being generated. In effect, the chat window, once a place to read answers, now serves as a place to use tools as well. OpenAI said it still needs to improve both the model’s judgment about interface design and the range of screens it can produce.
How much faster is the first response?
GPT-6 can start answering first and fill in the content as it goes. According to OpenAI, on questions that require a web search, GPT-6 Instant took an average of 44% less time to begin its response than GPT-5.6 Instant.
OpenAI also shared internal test results that measured speed and answer quality together. In a test of everyday tasks for AI agents — AI that handles work on its own — GPT-6 Extra High began responding in the same amount of time as GPT-5.6 Medium. Even so, the company said it posted a higher overall score than GPT-5.6 Extra High. On difficult questions, it added, the model addressed the heart of the question accurately more often than GPT-5.6.
What changed in the safeguards
OpenAI said the model is now more resistant to attempts to circumvent its safeguards, specifically tactics that stretch across multiple turns of conversation while shifting the method of attack. It also said it strengthened protections against high-risk misuse involving cyberattacks, biological threats, and violence. The company added that the model was trained to better recognize and communicate its own limits when it lacks the information or tools it needs.
When does it reach my plan?
The staged rollout begins over two days, depending on the plan. Plus, Pro, Business, and Enterprise start receiving it in the Chat tab on October 7. Free and Go follow on October 8. Whether Enterprise users can actually use it depends on how their company administrator has configured things.
The underlying model differs by plan as well. Plus, Pro, Business, and Enterprise get GPT-6 Sol, while Free and Go get GPT-6 Luna. Both models are tuned for everyday conversation. These changes apply to ChatGPT’s conversation interface. The models powering ChatGPT Work and Codex remain unchanged.
This update adds ready-to-use tools directly to ChatGPT’s answers and shortens the wait for a first response. For users in Korea, the key things to watch are when the change actually lands on their own plan and how the interface comes together in Korean-language conversations. How well the model builds the right tool for a given question is another thing worth watching going forward.