ChatGPT Returns to Service After Suffering Outage That Affected ChatGPT Work, Paid Users

ChatGPT experienced a partial outage on August 31, leaving users facing elevated errors and latency while ChatGPT Work became unavailable for some paid subscribers.

The disruption affected ChatGPT Work in particular, with users across multiple subscription plans unable to start new tasks or continue tasks already in progress. OpenAI confirmed the incident through its status page and later reported that the affected services had fully recovered.

ChatGPT Outage Causes Errors and High Latency

The cause of the ChatGPT service outage identified by OpenAI was elevated latency followed by elevated errors affecting ChatGPT Work. OpenAI initially reported increased latency at 3:04 p.m. ET before updating the incident to elevated errors 17 minutes later.

The company later said Plus users were particularly affected because Work mode was unavailable. The issue subsequently expanded to users across multiple subscription plans, who could be unable to start or continue ChatGPT Work tasks.

OpenAI has not publicly provided a deeper technical explanation for why the elevated latency and errors occurred. The company instead said it had identified the service problem and was working on a mitigation.

Paid ChatGPT Users Hit by ChatGPT Work Outage

The ChatGPT Work outage did not affect every user in the same way. OpenAI specifically noted that Plus subscribers were particularly impacted during the incident, while later updates said users across multiple subscription plans could experience problems.

Glitchwire reported that the disruption began around 11:04 a.m. ET and that Work mode was unavailable for some Plus subscribers. Users affected by the outage could encounter failures when attempting to launch new Work tasks or continue existing ones.

The incident therefore did not represent a complete shutdown of every ChatGPT function. Instead, the outage centered on ChatGPT Work, with errors and latency affecting the feature and some users' ability to use it.

OpenAI Applies a Mitigation

OpenAI continued updating the incident throughout the afternoon as engineers worked on a fix. At 8:01 p.m. ET, the company said it had applied a mitigation and was monitoring recovery.

At 8:28 p.m. ET, OpenAI marked the incident as resolved and stated that all impacted services had fully recovered.

BleepingComputer also reported on the disruption after OpenAI acknowledged the elevated errors and latency, noting that users across multiple subscription plans could be unable to start or continue ChatGPT Work tasks.

ChatGPT Is Back After the Service Outage

The ChatGPT outage was ultimately resolved after OpenAI deployed its mitigation. The incident lasted several hours and primarily affected ChatGPT Work, with Plus users among those most affected during the disruption.

OpenAI's status page now lists the incident as fully recovered, although individual availability can vary by subscription tier, model, and feature.

Originally published on Itech Post