OpenAI ChatGPT Outage History

OpenAI experienced a 13-hour incident where Codex context compaction was slower than expected, affecting both Codex and ChatGPT services. The issue was traced to a configuration error that caused performance degradation in Codex operations. The problem was resolved by applying a mitigation fix and monitoring the recovery.

Android ChatGPT business users experienced issues switching between workspaces, affecting login functionality and general ChatGPT access. The incident was classified as minor and lasted 11.6 hours while OpenAI investigated the problem. The service disruption specifically impacted the workspace switching feature for business users on the Android platform.

Users experienced errors when attempting to access chat support through OpenAI's help center, affecting both login functionality and ChatGPT services. The incident was identified and a mitigation was implemented to resolve the access issues. The service disruption lasted 4 minutes with users temporarily directed to email support as an alternative.

OpenAI ChatGPT experienced login issues specifically affecting FedRAMP users who had recently logged out, preventing them from accessing the service. The incident lasted 23.2 hours before a mitigation was applied and the system returned to normal operation. Both ChatGPT and the login system were impacted during this period.

OpenAI experienced an 18.3-hour incident where users encountered increased rate limiting issues with Codex and ChatGPT services. The company investigated the problem and applied mitigation measures to resolve the rate limiting issues. The incident was classified as minor and has been resolved with ongoing monitoring of the recovery.

ChatGPT 5.5 Thinking experienced elevated latency and error rates affecting both web and ChatGPT components for 1.3 hours. OpenAI identified the issue, implemented a mitigation, and monitored the recovery until service was restored.

OpenAI ChatGPT experienced elevated error rates affecting paid plan users for 1.5 hours, preventing messages from loading and generating responses with "error in message stream" notifications. Users reported being unable to get responses to their prompts despite being able to type messages, with the service consistently returning error messages instead of functioning normally. The issue was mitigated and the service returned to normal operation after monitoring confirmed recovery.

OpenAI ChatGPT experienced a major outage affecting logged-out users' ability to access the service, with login functionality also impacted for 24 minutes. Users reported content failing to load and unresponsive "Try again" buttons when attempting to use the service. The issue was resolved through applied mitigation measures and the service was restored to normal operation.

OpenAI's Codex 5.5 engines experienced elevated error rates affecting both ChatGPT and Codex services, with free users encountering high error rates and all Codex users experiencing degraded latency. The incident lasted 2 hours, during which the team identified the issue as specifically impacting gpt-5.5 performance. A mitigation was successfully applied and the service recovery was being monitored.

OpenAI ChatGPT experienced elevated errors affecting file uploads and Codex Cloud task creation for 4.2 hours. Users encountered difficulties uploading files to ChatGPT and creating tasks in Codex Cloud during this period. The issue was resolved after OpenAI applied a mitigation and monitored the recovery.