The first investigating update (Investigating) is displayed as Oct 7, 2026, 11:42 PM, saying the errors were identified and mitigation was under way. At 11:48 PM the incident moved to the monitoring stage (Monitoring), with mitigation applied and errors no longer observed; at 12:03 AM on Oct 8 OpenAI marked all impacted services recovered. The page does not label the timezone of these displayed timestamps or separately publish the actual incident start and recovery-completion times, so status-update times must not be treated as exact incident boundaries. OpenAI did not disclose the root cause, affected account count, impact percentage, or any user-specific remediation.
Service Incident and Recovery | FedRAMP / GPT-5.6 Instant
OpenAI Resolves Elevated GPT-5.6 Instant Errors in FedRAMP Workspaces
OpenAI’s status page reported elevated errors for GPT-5.6 Instant in FedRAMP workspaces and now marks the incident as recovered (Resolved). The listed affected component is FedRAMP; the notice does not announce a corresponding outage across general ChatGPT, Codex, or API services.
