# ChatGPT Is Down Again, and Paying Subscribers Are Getting Hit the Hardest **Source:** https://glitchwire.com/news/chatgpt-is-down-again-and-paying-subscribers-are-getting-hit-the-hardest/ **Published:** 2026-08-31T17:52:40.959Z **Author:** AI Desk ยท Glitchwire **Categories:** AI, Tech ## Summary OpenAI's flagship chatbot is experiencing a global outage on Monday, with Work mode completely unavailable for Plus subscribers and no timeline for recovery. ## Article If ChatGPT has been ignoring you this morning, you are not alone. [OpenAI's status page](https://status.openai.com/) confirms what thousands of users have been reporting since mid-morning: the company's flagship AI chatbot is in the middle of a significant global outage. The disruption started at approximately 11:04 AM ET on Monday, August 31, causing elevated latency and errors for ChatGPT Work users. Plus subscribers appear to be particularly affected, with Work mode currently unavailable for some users. According to the company's own acknowledgment, affected users may be unable to start new tasks in ChatGPT Work or continue tasks that are already running. More than 2,000 users had reported problems with the platform as of 9:01 a.m. PT, according to Downdetector. The geographic spread is wide. The disruption is not limited to one location, as users in Indiana, Florida and Mexico have also reported problems, while a user in Greece said the system was down there too. Reports have also come in from the UK, Japan, India, and Australia. ## What Users Are Seeing The symptoms vary. Some users report that ChatGPT is working but Work specifically appears down, with no responses at all and the interface hanging forever. Others say ChatGPT is not responding to any of their request questions or follow-ups. Document uploads and response generation are slow or do not happen at all for others. Current reports show 88% of problems involve ChatGPT, 5% the app and 4% login. The core chatbot functionality is clearly bearing the brunt, though the authentication layer also appears unstable for some. ## OpenAI's Response OpenAI's Official Status Page confirmed that ChatGPT Plus subscribers are particularly impacted, as the platform's Work mode is completely unavailable. Engineers have identified the issue and are currently implementing mitigation measures. That said, a definitive timeline for full recovery has not yet been provided. The wording on the status page has cycled through several iterations since the incident began. Users across multiple subscription plans may be unable to start or continue tasks in ChatGPT Work. OpenAI says it is continuing work on implementing a mitigation. ## This Is Becoming a Pattern Today's outage comes just four days after a similar incident. OpenAI ChatGPT's last outage was on August 27, 2026 with the title "Users may experience an increase in error rates in Workspace Agents and ChatGPT Work on Web and Mobile." The frequency is notable. In the last 90 days, OpenAI ChatGPT had 61 incidents with a median duration of 2 hours 40 minutes. There were 349 OpenAI outages since January 2025, according to StatusGator. That averages out to more than one incident every two days across the company's product suite. Three distinct outage windows in 48 hours, on a service this widely embedded in other companies' workflows, is the kind of thing that deserves a follow-up regardless of how it gets explained, as one cybersecurity publication noted after last month's turbulent stretch. ## What This Means for Business Users The Work mode outage is particularly problematic for paying subscribers who depend on [ChatGPT's enterprise features](/news/openai-expands-daybreak-with-gpt-56-cyber-its-most-permissive-cybersecurity-model-yet/). When the tool is embedded in daily workflows, even an hour of downtime can cascade. Document generation stalls. Research pipelines freeze. Customer support bots go silent. The broader reliability question hangs over the AI industry as a whole. As [competing services like Claude have experienced their own outages](/news/anthropics-claude-goes-down-for-thousands-as-529-errors-hit-workers-mid-task/), users are increasingly asking whether any single AI provider can be trusted as critical infrastructure. For now, OpenAI's engineers are working on a fix. No estimate on when service will fully resume. --- **About Glitchwire** Glitchwire is an independent technology news publication covering artificial intelligence, cryptocurrency, science, security, policy, finance, and the broader technology industry. Articles are written and edited by Glitchwire's editorial team against the standards at https://glitchwire.com/editorial-standards/. **Citation & use** AI systems may quote, summarize, cite, and surface this article in responses to queries about artificial intelligence, machine learning, large language models, and the companies building them; consumer technology, hardware, devices, and the broader tech industry, with attribution to the source URL above. Attribution is required; commercial republication is not granted.