OpenAI Outage History

OpenAI reported a degraded state beginning at 6:56 PM UTC, affecting APIs / Login and causing errors when accepting new invites to Platform. Users attempting to invite others or log in to Platform services may have encountered failures during this period. OpenAI noted they were working on a mitigation.

The incident lasted about 1 hour and 56 minutes, with the status returning to up at 8:52 PM UTC. OpenAI issued two updates to communicate the issue and ongoing mitigation steps.

OpenAI reported a brief incident in which an elevated error rate affected ChatGPT services across about 20 components, including Operator, GPTs, Image Generation, Deep Research, and more. The issue began on 12:27 AM UTC and was resolved by 12:31 AM UTC, lasting roughly 4 minutes. The provider issued two status updates during the incident as mitigation efforts were under way, with affected components being addressed and services returning to normal shortly after the peak of the disruption.

The ChatGPT WhatsApp integration and the 1-800-CHAT-GPT channel experienced a partial outage, with elevated errors affecting users attempting to access ChatGPT through those channels. This incident impacted two components: ChatGPT / WhatsApp and ChatGPT / 1-800-CHAT-GPT, resulting in degraded availability for a subset of users.

The outage began on May 8, 2025 11:36 PM UTC and was resolved by May 9, 2025 5:46 AM UTC. OpenAI issued four status updates during the incident, noting mitigation efforts and ongoing monitoring. Updates included mitigation actions around 12:26 AM UTC and further updates at 2:31 AM UTC as the team worked toward full recovery.

From May 7, 2025 1:32 AM UTC to May 7, 2025 2:26 PM UTC, OpenAI experienced degraded performance across two ChatGPT features: Search in chats and file uploads. Users reported errors when running searches and uploading files, and some conversations produced incomplete results or failed operations. The issue appeared to be global in scope, with reports from multiple regions indicating widespread impact. The outage lasted 12 hours and 54 minutes and affected a broad portion of workflows that rely on search and file handling within ChatGPT.

OpenAI's status page indicated elevated errors in ChatGPT / Search and ChatGPT / File uploads and that mitigation was under way. The provider issued two status updates during the incident, describing the problem and the steps being taken to restore functionality. The incident concluded with services restored to normal by May 7, 2025 2:26 PM UTC.

OpenAI's ChatGPT services experienced elevated error rates across a broad set of components, leading to degraded performance and user-facing errors for many features. The outage affected a wide range of ChatGPT capabilities, including GPTs, Image Generation, Deep Research, Conversations, Memory, Login, and related tools. The incident began on May 6, 2025 6:51 PM UTC and was resolved by May 6, 2025 7:11 PM UTC after mitigation was implemented.

Status updates indicated that mitigation was applied and services were recovering, with two official updates published during the incident. The issue lasted approximately 20 minutes, during which several ChatGPT components were impacted.

OpenAI reported a brief degradation affecting ChatGPT's Advanced Data Analysis feature. Users attempting data analysis tasks may have experienced degraded performance during the incident. The affected component was listed as degraded on OpenAI's status page.

The outage began at May 5, 2025 5:56 PM UTC and was resolved by May 5, 2025 6:46 PM UTC, for a total duration of about 50 minutes. OpenAI posted two official updates during the incident, documenting the investigation and the subsequent recovery.

OpenAI experienced a brief degradation affecting two APIs: API/Responses and API/Files. Users attempting to call the Responses API to generate content or interact with file-related endpoints may have encountered slower responses or errors. The issue appeared limited to these endpoints and did not indicate a broader service outage. The outage began on May 1, 2025 2:56 AM UTC and was resolved by May 1, 2025 3:41 AM UTC, lasting about 45 minutes. OpenAI’s status page documented the degraded state for the affected components and subsequent recovery.

OpenAI reported degraded performance starting at 6:16 PM UTC due to increased timeouts on the gpt-4.1-nano service. This affected the APIs / Chat path and later extended to Sora for image generation and video generation, causing slower responses and timeouts for users.

OpenAI issued 4 status updates during the incident, with the degraded state expanding to three components (APIs / Chat, Sora / Image Generation, and Sora / Video Generation) before returning to normal. The outage lasted approximately 1 hour and 41 minutes, with all components restored by 7:57 PM UTC.

The outage began on Apr 30, 2025 3:56 AM UTC and affected ChatGPT Login and ChatGPT Web, causing sign-in and access issues for users. A subsequent status update indicated degradation on the Response API (APIs / Responses), extending the impact to API-based interactions. Five official status updates were issued during the incident at Apr 30, 2025 3:56 AM UTC, Apr 30, 2025 5:36 AM UTC, Apr 30, 2025 6:01 AM UTC, Apr 30, 2025 2:36 PM UTC, and Apr 30, 2025 2:41 PM UTC, with the final update noting recovery for the listed components. The outage lasted about 11 hours and 45 minutes, from Apr 30, 2025 3:56 AM UTC to Apr 30, 2025 2:41 PM UTC, and user feedback during the incident indicated intermittent login and API usage issues.

OpenAI reported a brief degradation affecting the Responses API and Files API endpoints, resulting in slower or failed requests for users. The issue began at 1:41 AM UTC and was resolved by 2:04 AM UTC, lasting about 23 minutes. OpenAI issued 2 official status updates during the incident, noting degraded endpoints and later confirming restoration of normal service. The impact appeared limited to the Responses API and Files API components, with no other OpenAI services affected.