9:52 AM - 12:13 PM UTC (2h 22min duration)
Service degradation from Azure OpenAI causing failed assistant errors
The incident has been resolved.
Impact window: From 2:30 pm IST - 4:36 PM IST, some users experienced failed assistant errors.
Remediation: Errors from Azure OpenAI have subsided, and the system is operating normally.
We apologize for the disruption and appreciate your patience while we addressed this issue.
4:36 PM - 6:01 PM UTC (1h 25min duration)
Chat instability caused by hitting OpenAI capacity
OpenAI has increased our capacity. We longer see errors related to this. Please contact our support team if you face any issues
9:35 AM - 9:35 AM UTC (0min duration)
Chat/Assistant errors due to AzureOpenAI outage
Between 09:08 UTC on 16 Jul 2026 and 11:43 UTC on 16 Jul 2026, Azure experienced a platform issue affecting the Azure OpenAI Service service in the East US 2 region. During this incident, some customers experienced service interruptions with resources hosted in the region.
7:05 AM - 7:05 AM UTC (0min duration)
Elevated chat errors on AWS Bedrock customers
We are currently seeing elevated chat errors affecting our customers hosted on AWS due to availability issues with Amazon Bedrock. The impact is primarily affecting requests using Claude Opus 4.6 and Claude Sonnet 5.
Our team is actively monitoring the situation.
5:16 PM - Jul 13 1:40 PM UTC (2d 20h 24min duration)
Extension actions failing during auth and execution in Chat and Agents
A fix for the issue affecting extension actions has been deployed and is now resolved for nearly all customers.
We're reaching out directly to the small number of remaining customers who need to complete a few steps on their end.
If you had already authenticated to some of the Extension tools between 9th July 2026 and 13th July 2026 PST, then you will be requested to authenticate to those tools again.
3:31 AM - 5:55 AM UTC (2h 24min duration)
The incident has been resolved.
We apologize for the disruption and appreciate your patience while we addressed this issue.
9:26 AM - 3:02 PM UTC (5h 36min duration)
Client Token management API is failing
The incident has been resolved.
We apologize for the disruption and appreciate your patience while we addressed this issue.
No incidents reported