Azure Direct Models resource throttled due to CoTAbuse detection – Unable to create Azure Support Ticket

Santosh Korlepara 0 Reputation points
2026-07-21T08:11:53.1133333+00:00

Hello Microsoft Team,

We recently received an email stating that one of our Azure Direct Models resources has been throttled due to a potential violation of the Microsoft Online Services Acceptable Use Policy.

The email references:

Risk Category: CoTAbuse

Corrective Action: Throttled

Tracking ID: CQJ2-K1Z

Incident ID: 837981607

We have already started investigating our application logs and are reviewing recent requests for any prompt injection, jailbreak attempts, requests for chain-of-thought, system prompts, hidden instructions, or other prompts that may have triggered Microsoft's abuse detection. We are also implementing additional server-side safeguards to prevent similar requests from reaching the service.

Unfortunately, we are unable to create an Azure Support ticket through the Azure Portal/aka.ms/azsupt, so we are requesting assistance here.

Could a Microsoft moderator or Azure AI support engineer please help us with this incident?

Specifically, we would appreciate assistance with the following:

Reviewing Incident ID 837981607 and Tracking ID CQJ2-K1Z.

Providing any available diagnostic information that can help us identify the offending request in our application logs, such as:

Request ID

Trace ID

Correlation ID

Operation ID

UTC Timestamp

Deployment name (if applicable)

Region (if applicable)

Advising whether there are any additional remediation steps we should take before requesting removal of the throttling.

We understand that Microsoft may not be able to disclose the exact prompt or request content due to security policies. However, any diagnostic identifiers that can be correlated with our application logs would be extremely helpful.

If additional information about our Azure resource or subscription is required, we are happy to provide it through a private support channel or direct message.

Thank you for your assistance.

Content Safety in Foundry Control Plane
Content Safety in Foundry Control Plane

An Azure service that enables users to identify content that is potentially offensive, risky, or otherwise undesirable. Previously known as Azure Content Moderator.

0 comments No comments

Your answer

Answers can be marked as 'Accepted' by the question author and 'Recommended' by moderators, which helps users know the answer solved the author's problem.