The wall-of-text responses that ChatGPT is known for will apparently soon be a thing of the past. The company announced Wednesday that its GPT-6 family of models will now have access to “Intelligent UI,” which will result in the chatbot producing more visual outputs, including “graphics, tappable buttons, forms, charts, and interactive experiences.”

Per the company, when it makes sense for an answer to include more visualizations or interactive elements, the model will build those directly into its response—including some with functioning user interfaces. In an example of the Intelligent UI feature, OpenAI showed a design breakdown of a 7-speed bike, complete with a visualization of each part and a clickable button for each one that allows the user to read more about each element. Another example showed a plan for a lamb roast, including a menu with images of each dish (which appear to have been generated by OpenAI’s image generation model) and a clickable button to change the number of planned dinner guests to scale the ingredients needed for each dish.

OpenAI says ChatGPT will automatically generate these interactive interfaces when it deems it the most appropriate option, which it can pull from a library of components that the model was trained on. For more straightforward prompts, the chatbot will supposedly stick to text for simpler responses—but OpenAI noted that users can request a UI build whenever they want, and the chatbot will comply.

OpenAI’s “Intelligent UI” is reminiscent of Google’s “Generative Interface,” which it introduced with the launch of Gemini 3 last year .

Source link