AnswersPublished23 sources
ChatGPT Work Outage on August 31: What Happened and Who Was Affected
The August 31, 2026 incident was primarily a ChatGPT Work disruption that began around 11:04 a.m. OpenAI reported elevated errors and latency, with users unable to start or continue Work tasks across multiple subscription plans.
Published byEdited with GPT-5.6 LunaImages generated with GPT Image 1.5
Research answer
Key takeaways
- The August 31, 2026 incident was primarily a ChatGPT Work disruption that began around 11:04 a.m.
- OpenAI reported elevated errors and latency, with users unable to start or continue Work tasks across multiple subscription plans.
- Unlike the broader July outage, which also affected ChatGPT, the API, and Codex, the August incident was centered on ChatGPT Work; its cause was not publicly attributed to Codex growth or quota resets.
AI Prompt
openai.comCreate a landscape editorial hero image for this Studio Global article: What happened during OpenAI’s worldwide ChatGPT outage on Monday, August 31—including when it occurred, how many users reported problems on. Article summary: The August 31 disruption was a global ChatGPT Work incident, rather than a confirmed full-platform failure across every OpenAI product. It began at roughly 11:04 a.m. ET; OpenAI said it had identified elevated errors and. Topic tags: general, general web, user generated. Style: premium digital editorial illustration, source-backed research mood, clean composition, high detail, modern web publication hero. Use reference image context only for broad subject, composition, and topical grounding; do not copy the exact image. Avoid: logos, brand marks, copyrighted characters, real person likenesses, fake screenshots, UI text, readable text, watermarks, charts with fak
OpenAI’s August 31, 2026 incident was a ChatGPT Work outage, not a confirmed failure of every OpenAI product. The disruption began at approximately 11:04 a.m. ET, according to contemporaneous reporting, while OpenAI’s status updates described elevated errors and latency affecting Work users. 1
3
What happened during the August 31 outage?
The main failure involved ChatGPT Work tasks. OpenAI said users across multiple subscription plans could be unable to start new tasks or continue tasks already in progress. The status incident was classified as a partial outage while the company worked on a mitigation. 2
3
12
Reports described a mix of slow responses, errors, and disrupted Work workflows. The available evidence points to degraded functionality in ChatGPT Work rather than a simultaneous outage across ChatGPT, the API, and Codex.
OpenAI’s public wording identified the symptoms—“elevated errors and latency”—but did not provide a repair estimate in the incident updates reviewed for this report. The status history later recorded that the impacted services had fully recovered. 3
7
10
How many users reported problems?
Downdetector totals changed quickly depending on when the page was captured:
- One contemporaneous report cited more than 400 OpenAI problem reports shortly before 11:30 a.m. ET. About 80% of those reports concerned ChatGPT.
7
- Another snapshot recorded more than 2,000 ChatGPT reports by 9:01 a.m. PT.
4
These numbers should not be read as a count of affected customers. Downdetector records user-submitted problem reports, and the total can vary by region, timing, and the type of service being tracked.
Which plans and services were affected?
OpenAI said the incident could affect users on multiple subscription plans. Its status updates also indicated that Plus users were particularly impacted because Work mode was unavailable for some of them. 3
4
That does not prove that Plus was the only—or uniquely most—affected tier. Other reporting said users on different plans were also experiencing problems, so the strongest supported conclusion is that the outage was concentrated in ChatGPT Work but was not formally limited to one subscription level. 2
3
The distinction matters because Work was rolling out to paid plans other than Free and Go, including Pro, Pro Lite, Enterprise, Edu, Plus, and Business. 6 A Work-specific incident could therefore affect users differently depending on whether the feature was available in their account and workspace.
Did OpenAI give a resolution time or social-media explanation?
The available incident updates did not include a public resolution timetable. OpenAI said it had identified the elevated errors and latency and was implementing a mitigation, but the material reviewed here does not show a substantive social-media explanation or a publicly stated root cause during the outage. 3
7
OpenAI’s status history later showed that the impacted services had recovered. That confirms the eventual outcome, but it does not explain what caused the incident. 10
How it compares with the late-July outage
The August 31 incident was narrower than the late-July disruption described in the available reporting. On July 25, users reported login failures, missing or non-loading conversations, and access problems across the ChatGPT website and mobile app. The disruption also affected OpenAI’s API and Codex. 8
9
That difference is important:
- August 31: ChatGPT Work showed elevated errors and latency, with users unable to start or continue Work tasks.
- July 25: ChatGPT, the API, and Codex were all reported as affected, making it broader in product scope.
8
9
The two incidents fit a broader pattern of reported OpenAI service interruptions, but they should not be treated as the same failure or assigned the same cause.
What do outage trackers say about OpenAI’s reliability?
StatusGator’s outage-history page lists 377 OpenAI outages since January 2025. 49 That figure is a tracker’s historical count, not an official OpenAI reliability statistic, and the available evidence does not independently establish a precise total for the previous 90 days.
Other monitoring pages provide different views of service health and may count incidents, components, or user reports differently. For that reason, outage totals are best used as indicators of reported disruption rather than as a definitive measure of customer impact.
Did Codex growth or quota resets cause the outage?
There was a notable pattern of Codex and ChatGPT Work quota resets during June, July, and August. 32 Separate reporting also said Codex and ChatGPT Work had crossed 20 million active users in August.
39
Those developments may help explain why users were watching capacity and usage limits closely, but they do not establish that rapid Codex growth caused the August 31 outage. OpenAI’s incident updates attributed the event only to elevated errors and latency and did not publicly link it to Codex growth, quota resets, or infrastructure strain. 3
12
The defensible conclusion is narrower: rapid adoption and repeated quota-management changes provide context for the reliability concerns surrounding ChatGPT Work and Codex, but the cause of this specific outage remained unconfirmed in the available material.
Studio Global AI
Continue your research
This page includes a source-backed answer you can continue inside Studio Global.
People also ask
What is the short answer to "ChatGPT Work Outage on August 31: What Happened and Who Was Affected"?
The August 31, 2026 incident was primarily a ChatGPT Work disruption that began around 11:04 a.m.
What are the key points to validate first?
The August 31, 2026 incident was primarily a ChatGPT Work disruption that began around 11:04 a.m. OpenAI reported elevated errors and latency, with users unable to start or continue Work tasks across multiple subscription plans.
What should I do next in practice?
Unlike the broader July outage, which also affected ChatGPT, the API, and Codex, the August incident was centered on ChatGPT Work; its cause was not publicly attributed to Codex growth or quota resets.
Sources
glitchwire.comChatGPT Is Down Again, and Paying Subscribers Are Getting Hit the Hardest — Glitchwire
getmaxim.aiIs ChatGPT Down? OpenAI API & Platform Status | Bifrost
thetechoutlook.comChatGPT Work Faces Elevated Errors and Latency, OpenAI Working on a Fix - The Tech Outlook
status.openai.comOpenAI Status