Inside this story Context & sources

This article links to 2 sources. Source links are provided below so you can inspect the evidence yourself.

  • Interactive responses arrive in ChatGPT
  • Rollout depends on tier and workplace settings
  • Inspect the answer behind the interface

An answer you can use

OpenAI began rolling out GPT-6 with Intelligent UI on 7 October, giving ChatGPT a way to build interactive responses inside a conversation. Instead of receiving only paragraphs, users may see charts, buttons, forms or a small tool shaped around their question. [1]

The company says the global rollout started with Plus, Pro, Business and Enterprise, with Free and Go following from 8 October. Workplace administrators can affect Enterprise availability. The change applies to ChatGPT’s Chat experience; OpenAI says the models powering Work and Codex are unchanged by this release. [1]

Examples in the announcement include a bill splitter, interactive explanations and visual comparisons. These are OpenAI’s demonstrations, rather than independent evidence that every generated interface will work correctly. [1]

The conversation becomes a workspace

This shift has a useful point of comparison in Claude’s artifacts. Anthropic’s current help guide describes designs, documents, dashboards and small interactive tools that open beside a conversation, where people can edit, revisit and share them. Artifacts are available across its consumer and workplace plans, although particular templates and beta features have different availability. [2]

Our reading is that the more useful distinction is between an explanation for the moment and an object somebody will maintain. An interactive answer could help with the first; an editable document could suit the second. Either way, the interface should make the work easier to inspect and revise.

A sensible test for the new UI

For a South African reader, a useful trial might be a rand-denominated expense split or an interactive comparison built from information they provide. These are suggested experiments, not claims about locally verified accuracy or availability on every account.

The review should cover the result as well as the appearance. Does a changed input update the calculation? Are units and assumptions visible? Can another person follow the explanation without guessing how it works? A polished control panel still needs a correct underlying answer.

The interesting question is whether the interface saves a second task: rebuilding a table, drawing a diagram or opening a calculator. If it does, the chat becomes a more practical place to finish a piece of work.

Sources & further reading

1. GPT-6 and Intelligent UI for everyone

Vendor announcement and demos; no independent benchmark claims repeated.

2. What are artifacts and how do I use them?

Live help page says updated this week; context about another product, not corroboration of OpenAI rollout.

AI-assisted research and writing, with a separate source review. OpenAI’s rollout is reported from its announcement; Claude’s documentation supplies product context. The suggested local experiments are editorial analysis.

THE OPEN TABStay curious. Keep reading.