OpenAI announced on October 7, 2026 that it is moving ChatGPT’s Chat experience to the GPT-6 generation on every plan, including the free one. Last month the company introduced its first GPT-6 models for paying customers. Now it is bringing a new GPT-6 model to the more than 1.2 billion people who use ChatGPT each week. The headline feature is Intelligent UI: instead of answering only in text, ChatGPT can build a response with charts, maps, buttons, forms and even small tools you can use right in the conversation.

Screenshot of OpenAI's official blog post GPT-6 and Intelligent UI for everyone
Screenshot: OpenAI’s official blog

What Intelligent UI does

According to OpenAI, the model was trained to compose answers from text, visuals and interactive elements, and to decide how to combine them based on the question. A comparison might appear side by side, an explanation might come as a clickable diagram, and when plain text is the most useful answer, the model is supposed to choose text. OpenAI groups its examples into three kinds:

  • Everyday planning: a cooking timeline next to a recipe, or road trip stops shown on a map with notes.
  • Learning: interactive explanations of topics such as the central limit theorem or the Monty Hall problem, where you change an input and see what happens.
  • Tools on request: a savings calculator, a bill splitter for dinner with friends or a small game you play inside the chat.

Under the hood, OpenAI built a library of ready-made interface components and a compiler that processes the interface while the model generates it, so the answer appears progressively instead of all at once. The company itself says the model’s design judgment still needs work.

It starts answering before it finishes thinking

The second change is about waiting. Until now, reasoning models had to finish thinking before showing an answer. GPT-6 was trained to interleave thinking and writing: it starts building the answer from what it already knows and adds to it as it goes. In OpenAI’s internal testing, GPT-6 Instant started answering 44% sooner on average than GPT-5.6 Instant on questions that need a web search, and the company says it is also better at deciding when to search. These are OpenAI’s own numbers and haven’t been independently verified.

Safety

OpenAI says it updated the model’s safety training based on real-world use, with stronger protections against misuse involving cyberattacks, biological threats and violence. It says GPT-6 is harder to talk out of its safety training, especially in attempts that build up over several messages, while refusing fewer harmless requests. A system card with the test results was published alongside the launch.

Who gets it and when

The rollout is global. It started on October 7 for Plus, Pro, Business and Enterprise users in the Chat tab, and expands to Free and Go users the following day. Paid plans get answers from GPT-6 Sol, while Free and Go get GPT-6 Luna, the same model behind the Decisions API we covered recently. In Enterprise workspaces, availability depends on admin settings. The models behind Work and Codex are not changing in this update.

Why it matters

Until now, OpenAI’s newest generation was mostly for paying subscribers and developers, for example with GPT-6.1 Sol. This update brings it to most users and changes the shape of the answer itself. If Intelligent UI works as described, questions about a recipe, a trip or a budget will come back as something you can work with directly, not just read. It’s still worth double-checking the results, especially when numbers are involved. More updates in our AI News section.

Sources

  • OpenAI, official blog: GPT-6 and Intelligent UI for everyone, October 7, 2026, openai.com