OpenAI: GPT-6 brings interactive UI responses to all ChatGPT tiers
What happened
On October 7, 2026, OpenAI rolled out Intelligent UI to paid ChatGPT tiers — Plus, Pro, Business, and Enterprise. Free and Go tiers followed on October 8. The feature comes with GPT-6: GPT-6 Sol powers the paid tiers, GPT-6 Luna powers Free and Go.
With Intelligent UI, ChatGPT can now respond with interactive elements rather than text alone. Answers may include charts, forms, tappable buttons, interactive diagrams, or small functional tools such as a savings calculator, a bill splitter, or a game. The model decides which format to use based on the question. For simple queries, it still returns plain text.
Under the hood, OpenAI describes a library of native components and a compiler that renders the interface progressively as the model generates its response.
Why it matters for designers
Until now, a conversation with an AI assistant produced text. The user then took that text somewhere else — into a spreadsheet, a design file, a form — to act on it. Intelligent UI collapses that step: the model generates both the content and the interface at the same time.
OpenAI says it evaluates generated interfaces for clarity, usefulness, and completeness, and frames a long-term goal as software that “adapts to people.” That framing raises a question that will be relevant to product designers: as models take on more decisions about layout, controls, and information density, what role does design intent play in how those decisions get made?
For now, the immediate relevance is practical. Designers building on ChatGPT or the OpenAI API should test how Intelligent UI responds to their use cases — it may produce interfaces that are useful as-is or that conflict with an established design system. Either outcome is worth knowing about.