OpenAI has announced a new tool called Intelligent UI for the ChatGPT app, for generating visual elements as part of the artificial intelligence (AI) chatbot’s outputs. The update comes after the new GPT-6 models were introduced for paid subscribers last month, and aims to make complex learning topics simpler to explain. This will be an evolution for the ChatGPT app which will now be able to add visuals, tables or interactive elements within responses. This is an attempt to build an interface around queries that would have otherwise generated text responses, often revolving around complex tasks or concepts.
GPT-6 with Intelligent UI is now rolling out globally to ChatGPT Plus, Pro, Business and Enterprise subscribers, and will be available in the Chat tab within the ChatGPT interface. Tomorrow, the rollout will include ChatGPT Free and Go tiers. GPT-6 in ChatGPT is available as GPT-6 Sol for Plus, Pro, Business, and Enterprise subscriptions, and a lighter GPT-6 Luna model for Free and Go subscribers.
“We’ve trained GPT-6 to compose responses using text, visuals and interactive elements, choosing how they fit together based on your question. With Intelligent UI, responses can include graphics, tappable buttons, forms, charts, and interactive experiences you can use directly in your conversation,” the company says, in a statement. OpenAI says more than 1.2 billion people use ChatGPT each week.
They clarify that element selection will depend on the type of query or what a user is specifically asking for. “A comparison might work best side by side. An explanation might be easier to understand through an interactive diagram. And when a simple text answer is the most useful response, ChatGPT can give you that,” they say.
OpenAI says they’ve built a library of native, streamable components, along with a compiler that processes the interface as the model generates it. The library gives each response a familiar design foundation while leaving the model room to decide how the pieces fit together. The compiler allows the interface to appear progressively as the model generates it, without waiting for the entire response to be complete.
“We expanded our training methods to help the model make thoughtful decisions about content, layout, visuals, and interaction. This included evaluating the interfaces it creates for clarity, usefulness, and completeness. GPT‑6 learned to use the component library and make good design decisions, including how to organise information clearly, when to use interactivity, and when a simple text response is enough,” they explain. There is an admission that the model’s design judgement will improve with time.
The Frontier AI conversation comes to the fore with the GPT-6 generation models allowing for ChatGPT to begin answering a query while it is still thinking. “With the introduction of reasoning models, users could get help with much harder problems, but had to wait for the model to finish thinking before getting an answer. Now, ChatGPT can interleave thinking with answering,” the AI company says.
For this, training takes into account a user’s waiting time and progressively composes an answer using its knowledge and findings up to that point. It is trained to build answers by working with multiple partial responses, with each tasked to add useful information without what OpenAI calls unnecessary fillers. The attempt will be to keep the full answer as cohesive and factual as a response otherwise written all at once.
OpenAI claims that in an internal evaluation of high-value, everyday agentic tasks, GPT-6 Extra High is able to begin answering in the same amount of time as GPT-5.6 Medium while achieving a better overall score than GPT-5.6 Extra High.