OpenAI says GPT-6 can generate charts, forms and task-specific tools in answers. RuntimeWire's August 26th report documented interactive-answer infrastructure, without establishing the Intelligent UI branding or a GPT-6 connection.
By [Ryan Merket](https://runtimewire.com/author/ryan-merket)
· Published
· Updated
Primary source: [X / OpenAI](https://x.com/OpenAI/status/2107894997538525580)
Why it matters #
OpenAI is turning ChatGPT answers into small, task-specific interfaces. The rollout tests whether users will use answers that do things, rather than only explain, instead of moving to separate apps.
OpenAI began rolling out GPT-6 with Intelligent UI in ChatGPT on October 7th, adding interactive charts, forms and task-specific tools directly to answers. The feature is arriving in the Chat tab, where users can explore an explanation, use a generated calculator or interact with a visual without leaving the conversation, according to OpenAI's announcement on X and its product post.
[https://x.com/OpenAI/status/2107894997538525580](https://x.com/OpenAI/status/2107894997538525580)
For [Sam Altman (@sama)](https://x.com/sama), OpenAI's co-founder and CEO, the release extends a product strategy built around putting AI in front of a large consumer audience. Before OpenAI, Altman co-founded Loopt, a location-based social networking startup that [Green Dot acquired in 2012](https://ir.greendot.com/news-releases/news-release-details/green-dot-completes-acquisition-loopt/) for approximately $43 million in cash and retention-based incentives. OpenAI says more than 1.2 billion people use ChatGPT each week, giving this kind of product change a potential reach far beyond a standalone productivity app. That user figure is the company's own claim.
Intelligent UI changes what an answer can contain. GPT-6 can choose to combine text with graphics, charts, buttons, forms and interactive experiences, rather than returning only prose and links. OpenAI's examples include a road-trip plan shown on a map, an interactive explanation of a complex idea, a retirement savings calculator, a dinner bill splitter and a game. The choice of format is supposed to depend on the question; OpenAI says the model can still answer in plain text when an interface would add little.
The feature relies on software OpenAI built around the model. The company says it created a library of native, streamable interface components and a compiler that processes them as GPT-6 generates a response. That lets the interface appear progressively, rather than waiting for the model to finish composing the whole answer. OpenAI also says it trained the model to make decisions about content, layout and when to use interaction, and evaluated generated interfaces for clarity, usefulness and completeness.
RuntimeWire first uncovered OpenAI's underlying GenUI interface system on August 26th, six weeks before today's announcement. That investigation documented interactive components embedded in answers, individual widgets that could refresh after delivery, and 467 versioned "Learning Block" types. It did not establish the Intelligent UI branding, a GPT-6 connection or the October rollout.
That architecture puts the interface inside OpenAI's product rather than asking users to open a separate app for each small task. A savings calculator created in the exchange where someone asks about saving is immediately usable in that conversation. The release does not establish a new public interface-building product: OpenAI describes Intelligent UI as a ChatGPT capability, and says the update applies to the Chat tab. The models powering Work and Codex are unchanged.
Availability is split by plan. OpenAI says the global rollout starts October 7th for Plus, Pro, Business and Enterprise users, with Free and Go users scheduled to receive it beginning October 8th. Enterprise access depends on workplace administrator settings. GPT-6 Sol powers the listed paid tiers; GPT-6 Luna serves Free and Go. OpenAI describes both as tuned for everyday conversation. The models had already reached paid customers in other OpenAI products the previous month, making this release an expansion of GPT-6 into the main ChatGPT experience rather than the first arrival of the model family.
The business bet is that a conversational interface can handle more of the small jobs that currently send people to a calculator, map, explainer or simple app. OpenAI's announcement is focused on examples created and used within ChatGPT; it does not describe these generated interfaces as independent apps users can publish elsewhere. The initial rollout therefore tests whether people will use an answer that does something, not just one that explains something. The company's next challenge, by its own account, is improving the model's design judgment and expanding the kinds of interfaces it can create.