OpenAI Outage History

OpenAI reported an incident affecting ChatGPT / Conversations with elevated errors impacting all paid users. The issue began at Jul 21, 2025 12:41 PM UTC and lasted for about 1 hour 35 minutes, with restoration by Jul 21, 2025 2:16 PM UTC. User reports started around 12:15 PM UTC and included at least one Berlin user noting “doesnt work” at 12:58 PM UTC. The impact appeared global for paid accounts, with some users experiencing timeouts and degraded responses. OpenAI’s status page described the issue as elevated errors on ChatGPT for all paid users and indicated mitigation was underway, followed by a restoration announcement. Two official status updates were issued during the incident, as the ChatGPT / Conversations component transitioned from degraded to normal.

OpenAI experienced a brief outage affecting the image processing features in ChatGPT Vision for paid users. Users attempting image-related capabilities reported elevated errors during the incident, with the impact appearing limited to the Vision path. The outage lasted about 25 minutes, beginning at 8:51 AM UTC and ending at 9:16 AM UTC.

OpenAI issued 2 status updates during the incident, acknowledging the elevated errors and describing mitigation efforts as they worked to restore normal service.

OpenAI experienced a brief outage affecting the ChatGPT / Web component, specifically the file upload feature, with users unable to upload files in the ChatGPT Web interface. The outage began on 11:06 PM UTC and was resolved by 11:36 PM UTC, lasting about 30 minutes. OpenAI issued 2 status updates during the incident, outlining the issue and recovery.

OpenAI's ChatGPT experienced elevated network errors affecting users across all major clients: ChatGPT Desktop Windows, ChatGPT Desktop macOS, ChatGPT Web, ChatGPT Mobile iOS, and ChatGPT Mobile Android. The issue began at Jul 17, 2025 10:41 PM UTC and produced a degraded experience with connection errors for several hours. The outage ended at Jul 18, 2025 2:56 AM UTC, lasting about 4 hours and 15 minutes.

OpenAI issued two official status updates during the incident, outlining the investigation and restoration steps. The incident impacted five components and gradually returned to normal as stabilization occurred. A small volume of user reports surfaced, including at least one report from Chattanooga, Tennessee describing repeated network errors.

OpenAI APIs experienced elevated error rates and latency across a broad set of services, impacting user workflows such as authentication, content generation, and chat. The incident affected multiple components across the API surface, including API endpoints for Responses, Completions (legacy), Login, Images, Chat, and Uploads, with related Sora and image-generation features also impacted. The outage began when the OpenAI status page reported a degraded state at Jul 17, 2025 9:06 AM UTC and expanded over several hours as additional surfaces were affected. The outage lasted about 8 hours and 15 minutes and was resolved with a final recovery update at Jul 17, 2025 5:21 PM UTC.

OpenAI experienced a brief API outage affecting EU Data Residency customers. The outage began at 11:17 PM UTC and recovery began at 11:24 PM UTC. In total, about 15 API components were affected, including Chat, Completions, Embeddings, and Login, causing temporary unavailability or degraded performance for regional users.

OpenAI's status page documented the incident with updates that confirmed restoration and return to normal operations by 11:51 PM UTC.

The OpenAI outage affected Single Sign-On for ChatGPT and the Platform login APIs, causing login attempts to fail for a subset of users. The incident began at Jul 16, 2025 9:11 PM UTC and lasted about 5 minutes, with services restored by Jul 16, 2025 9:16 PM UTC. The affected components were ChatGPT / Single Sign-On and APIs / Login.

OpenAI posted two status updates during the incident describing the impact and mitigation. User reports during the event were limited, consistent with a brief disruption restricted to sign-in flows.

OpenAI experienced broad elevated error rates across multiple services, notably Sora Video generation and a wide range of ChatGPT features. Users reported server errors, timeouts, sign-in problems, and difficulty loading conversations or histories. The disruption appeared global and persisted for roughly two hours, beginning around Jul 16, 2025 1:36 AM UTC.

OpenAI's status page documented the investigation and mitigation across several updates. The provider issued seven updates between Jul 16, 2025 1:36 AM UTC and Jul 16, 2025 3:51 AM UTC, noting mitigation steps and ongoing monitoring as recovery progressed. By the final updates, many components had returned to normal, with recovery completed for the majority of services by around Jul 16, 2025 3:51 AM UTC.

Scope and impact: The incident affected more than a dozen ChatGPT subsystems in addition to Sora Video generation, including Web, Codex, Context connectors, Desktop and Mobile clients, Memory, File uploads, Dall-E, Login, Conversations, and related features. User reports tracked a global pattern of issues across regions, consistent with a widespread outage rather than a regional fault. Overall, the event represented a major disruption to OpenAI services during the window.

OpenAI reported elevated error rates across a broad range of ChatGPT features and API endpoints on 2025-07-10. The incident affected many components across ChatGPT user-facing experiences (ChatGPT features such as file uploads, single sign-on, WhatsApp integration, 1-800-CHAT-GPT, search, advanced data analysis, context connectors, DALL·E, GPT vision, voice mode, memory, conversations, GPTs, image generation, deep research, operator, Codex, desktop clients for Windows and macOS, login, and Web and mobile apps) as well as a large portion of OpenAI APIs (including Realtime, Responses, Login, Images, Embeddings, Batch, Fine-tuning, Files, Assistants, Audio, Completions, Chat, Compliance, Moderations, Uploads). The breadth of components impacted indicates a broad degradation of service rather than a single feature outage. The incident began at Jul 10, 2025 5:02 PM UTC and was resolved by Jul 10, 2025 7:46 PM UTC, lasting roughly 2 hours and 44 minutes. During the incident, OpenAI published three status updates documenting progress as teams investigated and then restored services. User reports remained minimal, with little external indication beyond the provider's notices.

OpenAI experienced elevated errors across the 4.1-mini API, causing degraded performance for developers and applications. A broad set of OpenAI API components were affected, including Assistants, Audio, Images, Fine-tuning, Realtime, Responses, Login, Batch, Embeddings, Files, Completions (legacy), Chat, Compliance, Moderations, and Uploads. The outage began at 9:31 PM UTC and was resolved by 9:57 PM UTC.

OpenAI published two status updates during the incident detailing a mitigation and ongoing recovery as components returned to normal. The incident touched a wide user base and was reflected in user-reported issues during the event, before a return to normal operation.