AIToday
Large Language ModelsAI Business & IndustryASCII.jpPublished: Nov 20, 2026, 16:00 JST

OpenAI users report Codex, ChatGPT Work quotas draining fast

Send on LINE
OpenAI users report Codex, ChatGPT Work quotas draining fast

3 Key Points

  1. What happened

    Since late October, users of OpenAI's paid plans have reported Codex and ChatGPT Work burning through usage quotas abnormally fast; one ChatGPT Plus user said a Work task hit the limit in 4 minutes 21 seconds with 100% remaining.

  2. Why it matters

    The two tools draw on a shared quota, so faster consumption means paying users may be blocked from starting new tasks for a period — eating into the work they assumed their plan covered.

WHO IT HITSIndividual paid ChatGPT subscribers, particularly developers and knowledge workers who run Codex and Work tasks daily, could find their quota exhausted mid-task and be unable to start new work until the period resets.

Not sure about something? Ask the AI

Questions and answers are published on this page.

Summaries like this, every morning.

Context & Analysis

The reports arrive against a backdrop of OpenAI already having admitted a similar quota problem with Codex. Users were told that a June 2026 case stemmed from mistaken usage restrictions on some accounts. That precedent gives the new complaints a familiar shape, even though the late-October reports remain unexplained.

What makes the shared-quota design consequential is its coupling: Codex, which hands programming work to the AI, and ChatGPT Work, which handles research and document preparation, draw on the same allowance. A surge in consumption in one therefore affects access to the other. Users have floated several possible causes — repeated reads of external tools, reprocessing of past conversations, or changes in the volume and type of data the AI handles — but these are speculation rather than confirmed findings.

For paid subscribers, the practical question is whether this is a display problem or a real increase in consumption. No statement has been confirmed from OpenAI on the late-October wave, so that distinction remains open.

FAQ
How much faster is the quota draining, according to users?
One user said work that once took about 5 hours and used roughly 30% of the quota now consumes it in about 3 hours — about 5.6 times the previous rate.
What is causing the faster consumption?
The cause is not clear. Users speculate the AI may be repeatedly reading external tool information or reprocessing past conversations, pushing usage beyond expectations.
Has OpenAI acknowledged this problem?
OpenAI has previously acknowledged that Codex quotas drained faster than expected, attributing a 2026年6月 case to mistaken restrictions on some accounts. No statement on the late-October reports has been confirmed.
Send on LINE

AI news that matters for your work, delivered every morning.

Pick your industry and the AI tools you use, and get news related to your work every day.

Free · 30 seconds with Google · unsubscribe anytimeWhat is AIToday? →

Ask AI

Ask AI anything about this article. The AI reads this article, earlier AIToday articles, and Wikipedia, and cites its sources. Q&As are published on this page for other readers too.

Questions and answers are published on this page.

Related Articles

Next articleWenge AI adds Google and Apple sign-in to ScienceOra