There were 262 OpenAI outages since January 2025 which are summarized below, including incident details, duration, and resolution information.
OpenAI not working for you?
OpenAI not working for you?
Minor
: OpenAI partial outages: SFT delays and web chat issues
Detected by StatusGator:
Officially acknowledged:
Outage ended:
(1 day, 23 hours and 30 minutes)
OpenAI experienced partial outages beginning at with delays in supervised fine-tuning (SFT) that affected the APIs / Fine-tuning workflow. SFT processing slowed and caused workflow delays for users. A separate disruption began around that briefly impacted web chat functionality for ChatGPT / Conversations, with some users reporting login problems and chats not loading. The combined incident caused degraded performance across SFT processing and web chat services, with recovery progressing over the following hours and the SFT component returning to normal by .
Minor
: OpenAI services degraded across multiple APIs
Detected by StatusGator:
Officially acknowledged:
Outage ended:
(4 days, 5 hours and 55 minutes)
OpenAI experienced broad degradation across multiple core APIs and ChatGPT features starting on with elevated error rates affecting connectors and related services. End users reported slower responses, intermittent errors, and difficulties with content generation, image generation, and login flows across components including APIs / Responses, APIs / Realtime, APIs / Images, ChatGPT / Image Generation, ChatGPT / Deep Research, ChatGPT / Agent, ChatGPT / Codex, Sora / Image Generation, and ChatGPT / Login. The issue progressed through and , with OpenAI noting ongoing service degradation and recovery attempts for the affected components.
Between and there was a clearly defined partial outage of the Responses API, impacting a subset of requests and error handling. The incident continued with multiple updates as services gradually recovered, culminating in most OpenAI services returning to normal availability by .
OpenAI published seven status updates during the incident, detailing investigations, mitigations, and the gradual restoration of functionality. The overall impact was significant in terms of degraded performance and partial outages across numerous components, but there was no full service outage for the majority of users.
Major
: OpenAI outage disrupts login and GPT features
Detected by StatusGator:
Officially acknowledged:
Outage ended:
(5 hours and 5 minutes)
OpenAI experienced a partial outage beginning at UTC, impacting 11 components across GPT-related services. Users faced login failures and widespread feature degradation, with services such as ChatGPT / GPTs, ChatGPT / Agent, ChatGPT / Codex, ChatGPT / Deep Research, ChatGPT / Search, ChatGPT / Memory, and related login and context components affected. The incident disrupted access to multiple GPT capabilities and authentication-related functions for several hours.
OpenAI ChatGPT experienced a brief outage affecting Android users attempting to sign up with Google on the mobile app. Sign-up attempts were failing or degraded, preventing new accounts from being created via Google on Android. The incident began at and was resolved by , lasting approximately 9 hours and 15 minutes. OpenAI issued 2 status updates during the incident, noting an active investigation and the eventual restoration of the Google sign-up flow in the ChatGPT Mobile - Android component.
Minor
: OpenAI API briefly degraded; chat completions and responses affected
Detected by StatusGator:
Officially acknowledged:
Outage ended:
(10 minutes)
OpenAI experienced a brief degraded performance affecting the APIs used for chat completions and responses. The incident began at with elevated error rates for gpt-5-nano-2025-08-07-4292, impacting two components: APIs / Chat Completions and APIs / Responses. The degraded period lasted about 10 minutes and was resolved by when service levels returned to normal. OpenAI posted two status updates during the incident. The updates acknowledged elevated error rates and noted that mitigation was underway, followed by confirmation that the affected endpoints had recovered.
Major
: OpenAI ChatGPT and Compliance API degraded for ~24 hours
Detected by StatusGator:
Officially acknowledged:
Outage ended:
(1 day and 10 minutes)
OpenAI experienced a global degradation impacting ChatGPT Conversations and the Compliance API ingestion pipeline, resulting in delayed data processing and user-facing errors. The incident began at , when OpenAI flagged Compliance API ingestion delays and marked the ChatGPT / Compliance API as degraded. Affected users reported slow responses and intermittent failures across ChatGPT features.
By , OpenAI indicated an increased error rate for ChatGPT Conversations, signaling ongoing impact across services. A mitigation was applied and recovery monitoring started around . The provider continued updates and, by , the ChatGPT / Compliance API had returned to normal.
The incident spanned approximately 24 hours and featured four official status updates documenting the progression from ingestion delays to broader errors and eventual restoration. Reports from users worldwide indicated widespread impact across regions; no single geographic pattern was observed beyond global usage. Recovery concluded with services restored to normal operation.
Minor
: OpenAI fine-tuning API latency degraded for ~21 hours
Detected by StatusGator:
Officially acknowledged:
Outage ended:
(21 hours and 35 minutes)
OpenAI experienced degraded performance in its Fine-tuning API, with fine-tuning jobs taking longer than expected due to high load. The issue began at and was resolved by , lasting about 21 hours and 35 minutes. Users reported slow or unresponsive fine-tuning workflows during this period. The incident affected the APIs / Fine-tuning component, and OpenAI posted two status updates during the event to communicate mitigation efforts and eventual recovery.
Minor
: OpenAI API Code Interpreter errors briefly degrade responses
Detected by StatusGator:
Officially acknowledged:
Outage ended:
(1 hour and 15 minutes)
OpenAI reported an incident affecting the Code Interpreter component of the OpenAI API. The outage began at when the Code Interpreter API area moved to a degraded state, causing elevated errors for users relying on Code Interpreter features. Some tasks that generate or download results were impacted. The outage lasted approximately 1 hour 15 minutes and was resolved by .
OpenAI published two official updates during the incident. The first, at , acknowledged elevated errors and noted mitigation efforts for the APIs / Responses component. The second, at , confirmed the issue had been resolved and services returned to normal. User reports spiked around UTC, with observations of API errors and problems generating or downloading results.
The OpenAI ChatGPT WhatsApp integration suffered a widespread outage, impacting messaging across multiple channels. The incident began on and was resolved by , lasting approximately 21 hours and 29 minutes. Eight components were affected, including ChatGPT / WhatsApp, ChatGPT / 1-800-CHAT-GPT, Login, Web, Mobile - iOS, Mobile - Android, ChatGPT / Conversations, and ChatGPT / Context connectors for Enterprise. Users across regions reported being unable to send messages or receive responses, indicating a global footprint across North America, Europe, and Asia.
The ChatGPT Context connectors component in OpenAI experienced a brief degraded performance starting at with increased errors affecting connector usage. The issue persisted for about 1 hour and 5 minutes and was resolved by . OpenAI issued 2 status updates during the incident documenting the investigation and restoration of normal functionality for the ChatGPT / Context connectors.
