Search fully restored in ChatGPT Work, Dots and thinking experiences following OpenAI outage

OpenAI has resolved the elevated errors affecting search functionality in its ChatGPT Work and Dots products. The company’s status page confirms that all affected services are now operating normally following the disruption.

The evidence

According to an official advisory from the OpenAI status dashboard, the incident involved elevated errors with search in Dots, ChatGPT Work and certain ChatGPT thinking experiences.reported The status page states that the issue is resolved and all affected services are operating normally again. OpenAI apologised for the disruption and thanked users for their patience during the investigation period. The advisory notes that availability metrics are reported at an aggregate level across all tiers, models and error types.reported It also clarifies that individual customer availability may vary depending on subscription tier, specific model and API features in use.

The decision

For enterprise IT buyers monitoring service reliability, this resolution signals a return to baseline operations for these specific interfaces. The fact that availability metrics are aggregated means that while the overall service is marked operational, individual teams might have experienced varying levels of access depending on their specific configuration. The advisory does not provide a root cause analysis or detailed timeline of the outage. It simply confirms the current operational status. This lack of granular detail suggests that while the immediate technical fault is cleared, deeper insights into the cause may not be publicly available in this specific status update.

The takeaway

The documented tradeoff here is between aggregate reporting and individual experience. While OpenAI declares the services operational, the variance in individual availability based on subscription and model usage means that not all users may have experienced the same recovery trajectory. The incident affected search capabilities specifically, leaving other core functions potentially untouched. This highlights the importance of monitoring specific feature sets rather than just overall service health. The resolution confirms that the search component has been restored, but the aggregate nature of the metrics limits precise verification for every single user instance.

Building on this

Developers integrating with ChatGPT Work or Dots should note that search functionality is a distinct component subject to separate availability metrics. The status page indicates that individual customer availability varies by subscription tier and specific model. This suggests that applications relying on search features should implement robust error handling and fallback mechanisms, as aggregate operational status does not guarantee uniform availability for every API call or user session. The fact that certain ChatGPT thinking experiences were also affected implies that complex reasoning or search-dependent workflows may be more susceptible to such disruptions. Developers should monitor the OpenAI status page for real-time updates, as the current resolution confirms normal operation but does not preclude future intermittent issues. The variance in availability based on API features in use means that testing should cover the specific models and tiers deployed in production environments.