AI Incidents
Back to incident register
Source reviewedOperational incidentMedium

OpenAI Responses API experienced elevated latency on the US endpoint

OpenAI's status page documented elevated latency affecting some Responses API requests on the US endpoint from August 31 into September 1 before mitigations restored service.

First observed
Sep 1, 2026, 12:27:56 AM
Disclosed
Sep 1, 2026, 12:27:56 AM
Status
Resolved
Confidence
99%
Organization
OpenAI
Last reviewed
Sep 4, 2026

Observed

Facts supported by sources

  • OpenAI reported elevated latency affecting some Responses API requests on the US endpoint.
  • The incident was first posted at 22:27 UTC on August 31 and marked fully recovered at 19:05 UTC on September 1 after mitigations and monitoring.

Assessment

Interpretation, not observation

  • OpenAI did not publish a root cause on the status page reviewed for this run.

Impact

Impact

Slower Responses API calls for affected US-endpoint customers over a multi-hour period.

Response

Response

OpenAI identified the issue, applied multiple mitigations, monitored recovery and then reported full service restoration.

Methodology: Facts, interpretations, and uncertainty are kept separate. Confidence describes the strength of the evidence, not a probability estimate.