OpenAI Outage History

OpenAI experienced an issue affecting customers using the Java SDK with Responses API streaming functionality. The incident impacted the Responses and APIs components for approximately 1.7 hours. The issue was mitigated and the service was restored to normal operation.

OpenAI experienced a major incident where ChatGPT FedRAMP workspaces failed to load, causing elevated errors for users trying to access these services. The issue lasted 23 minutes, during which OpenAI identified the problem and worked on implementing a mitigation. The incident has been resolved.

OpenAI experienced account creation issues affecting ChatGPT, APIs, Sora, and login services for 22 minutes. Users were unable to create new accounts across these platforms during the incident. The issue was identified and a mitigation was implemented to resolve the problem.

OpenAI experienced elevated 401 authentication errors across Chat, Responses, APIs, and Embeddings services for 36 hours due to timeout issues. The errors were intermittent and affected API endpoint access, preventing users from properly authenticating their requests. The issue was resolved through multiple mitigation deployments that gradually reduced error rates back to normal levels.

OpenAI experienced a minor incident where some users had difficulty accessing ChatGPT for 48 minutes. The company investigated the issue and applied a mitigation to resolve the access problems. The service was restored and monitored for stability.

OpenAI experienced a minor service incident affecting their API platform login functionality and account creation processes. The issue impacted users' ability to access the login system and APIs for 11 minutes. The incident was resolved after investigation by the OpenAI team.

OpenAI experienced issues with share link creation functionality on their web UI for 13 minutes on April 10, 2026. The service classified this as a minor incident affecting users' ability to create shareable links through the web interface. The issue was under investigation during the outage period.

OpenAI experienced elevated login errors affecting ChatGPT and APIs for 25 minutes. Users were unable to log into the impacted services during this period. The issue was identified and a mitigation was applied, with the service returning to normal operation.

OpenAI experienced a minor service incident where users accessing ChatGPT through VPN connections encountered access issues. The incident affected multiple components including the web interface, APIs, login, voice mode, image generation, and other core features. The issue was resolved after 1.4 hours when OpenAI applied a mitigation and confirmed recovery.

The OpenAI GET /v1/responses endpoint experienced an outage that prevented it from serving requests, affecting the Responses component and APIs. The incident lasted 35 minutes before the service was mostly restored. OpenAI investigated and mitigated the issue while monitoring the recovery process.