OpenAI plans to bring ChatGPT’s Chat and Work tabs together from December, while drawing ChatGPT and Codex usage from a shared allowance for Plus, Pro and Business plans. The change is intended to make it easier to move from asking a question to getting work done, but it also changes how some subscribers’ usage is counted.
OpenAI Watch analysis
What happened
ZDNET reports that the combined experience will first be tested with select users before a wider rollout. Codex will remain a separate experience in the ChatGPT desktop app, and OpenAI expects most users not to be affected by the transition.
Under the new arrangement, ChatGPT and Codex will use the same token allowance. OpenAI says usage depends on the model and task, with more advanced models and complex tasks generally using more. Instant chat will not draw from the allowance. Some users, depending on their usage and model choice, may have to wait for resets more often than they do now.
Why it matters
This is more than a tidier set of tabs. Putting chat and work in one place could make it easier to take an idea through to a website, presentation or web task, while the shared allowance makes the boundary between ChatGPT and Codex matter to subscribers’ usage.
The practical wrinkle is that the effect will vary by how people use the services. OpenAI expects most users’ experience to remain steady, but heavier users may need to keep an eye on how their work draws from the combined pool.
Our read
A simpler interface is welcome if it really reduces the hop between “help me think” and “help me do”. The allowance change is the part to watch: convenience is less charming if it quietly means reaching a reset sooner. Subscribers should check how their plan’s usage rules apply once the new experience reaches them.
What to watch
- When select-user testing begins and when the wider rollout follows.
- Whether OpenAI publishes clearer usage details for the combined allowance.
- Whether users on Plus, Pro and Business notice more frequent waits for resets.
Discussion spark: Should ChatGPT and Codex share one usage allowance for a smoother workflow, or should subscribers keep separate limits for each?
Sources and evidence
- Source update (2 November 2026, 18:01 UTC)
OpenAI Watch is independently operated by WittyWires. It is not affiliated with, endorsed by, or operated by OpenAI.