Capacity becomes a competitive issue
TechCrunch reported on September 10 that OpenAI paused new Pro sign ups after heavy use of Astra, the company’s recently launched model, placed unusual pressure on its infrastructure. Thibault Sottiaux, OpenAI’s product leader for core products including ChatGPT and Codex, said the Pro tier puts the greatest load on the company’s systems.
The company has described the pause as a targeted measure. Lower priced plans, API access and existing Pro subscriptions remain available. OpenAI has not said how long the restriction will last, how many potential customers were affected or which specific part of its infrastructure reached its limit.
That lack of detail matters because the decision is more than a temporary service notice. It suggests that premium pricing does not necessarily guarantee unlimited access to premium capability. In this case, the customers most willing to pay are also the customers likely to generate the most expensive workloads.
Astra, launched September 3, is built around reasoning, coding and computer use. Those tasks require more than producing a short answer. A model may need to inspect files, operate software, call tools repeatedly and revise its work over an extended session. The difference is similar to running a calculator versus hiring an analyst to work through an entire business process. The second service consumes far more time and capacity.
A test for the subscription model
The timing indicates that demand may have risen sharply after Astra’s release. OpenAI increased Codex usage limits as recently as the previous month, suggesting the present strain could be linked to adoption rather than a permanent shortage. The company’s ambitious claims about Astra, including its description of the model as the beginning of an “AGI era,” may also have intensified interest.
The episode creates difficult choices for OpenAI and its rivals. Providers can impose stricter quotas, route simpler requests to smaller models, introduce waiting lists or raise prices. Each option, however, risks weakening the promise that a subscription provides dependable access to the best available system.
For enterprise buyers, the lesson is broader. Model quality is only one procurement concern. Capacity, usage limits and the provider’s willingness to disclose constraints may become equally important. As AI systems shift from answering questions to performing extended tasks, infrastructure reliability will increasingly shape the competitive landscape.
Alex Carter is not a person. No notebook, no deadlines, no face behind the name — just a byline this newsroom publishes under. Here is the production line underneath it, because a name beside a portrait reads like a journalist, and this one is not one.
The models. Writing: gpt-5.6-luna and qwen3-max. Out on the live web: gpt-5.6-luna and gpt-5.6-terra. Pictures: gpt-image-1 and gpt-image-1-mini. Swap one in the newsroom and this line swaps with it — it is read off the machines, not typed here.
How a story is made
- Research. The searching model reads around the story, pointed at primary sources — the filing, the post, the repository — rather than at somebody else's write-up of them.
- Writing. The writing model drafts it against what was found, at Alex Carter's usual length and in Alex Carter's usual register.
- The loop. A reviewer reads the draft and sends it back with notes. Then reads it again. A piece can go round several times before it leaves the building.
- Enrichment. A quotation has to appear word for word on the page it is taken from. A chart may only use figures that appear in the source it cites. Whatever fails is dropped, and the reason is kept.
- Fact check. A last pass hunts for claims the article makes and its sources do not.
- A human stop. Sensitive subjects are held for a person to read before publication, and a person can kill any of it at any point.
If that sounds less like a newsroom and more like a factory: quite. It is called Press Factory.
This article was generated using AI and published automatically without human pre-publication review.
How this article was made
The article was produced by the Grandmonts Media News Engine using automated research, drafting and verification workflows. No human editor reviewed the article before publication. Grandmonts Media remains responsible for the published content. Errors can be reported at office@grandmonts.cz.