OpenAI has clarified the scope of the resolved API outage: all 14 components were affected, not just Chat Completions and Responses
OpenAI now states that the November incident with elevated error rates affected its entire API infrastructure, including Login, Agents, Embeddings, Realtime, and other services, not just Chat Completions and Responses as previously reported. The outage has already been fully resolved.
OpenAI has updated the status of the incident titled "Elevated Errors On API" and announced that it has been resolved — according to the status page, all affected services have returned to normal operation. The newly published list of affected components is, however, significantly broader than previously stated: it includes Batch, Fine-tuning, Files, Moderations, Responses, Decisions, Agents, Live, Images, Chat Completions, Audio, Realtime, Embeddings, and Login. According to the company, all of these components are now back in "Operational" mode.
An earlier report spoke of an incident with elevated error rates specifically for the GPT-5.6 Sol model API, listing only Chat Completions and Responses as the affected components. The sources therefore differ in the scope of affected services — one describes a narrowly defined issue affecting two components tied to a specific model, the other a broader outage across OpenAI's entire API infrastructure.
OpenAI has classified the incident as a technical/operational error, not a security issue. The exact time span of the outage and its cause are not stated in the sources.
Why it matters
For companies and developers building products on the OpenAI API, it matters that the outage apparently also affected authentication (Login) and other key components, not just partial functionality tied to a single model — which increases the real-world impact on the availability of downstream services. See the source article for details.
What was added since the original report
Verified updates
-
The incident affected all 14 components of the OpenAI API (not just Chat Completions and Responses); the outage included the Agents, Fine-tuning, Batch, Embeddings, Realtime, Audio, Images, and Login services, which the existing event does not mention; the incident was explicitly confirmed as a technical/operational error, not a security incident; the source is the official OpenAI Status page with an authoritative overview of the incident; all components individually confirmed as Operational in the new report
- The incident affected all 14 components of the OpenAI API (not just Chat Completions and Responses)
- The outage included the Agents, Fine-tuning, Batch, Embeddings, Realtime, Audio, Images, and Login services, which the existing event does not mention
- The incident was explicitly confirmed as a technical/operational error, not a security incident
- The source is the official OpenAI Status page with an authoritative overview of the incident
- All components individually confirmed as Operational in the new report
Relevant practical impact
What this means
For a business
Companies dependent on the OpenAI API faced, during the incident, a risk of outage across a wider set of services (including login) than originally announced two endpoints — it's worth checking your own integration logs from that period and considering status page monitoring or a backup solution for critical operations.
Risks and complianceCheck the original
Event sources
clearly official source · 1 publisher, 0 independent. We count feeds from the same owner only once.