Error 715-123420 - Backend escalation has had no update for over one month, requesting another Microsoft moderator to assist

保元 駿 0 Reputation points
2026-09-21T01:59:30.7833333+00:00

Hello,

I am requesting assistance with an Azure AI Foundry deployment issue that has already been investigated through another Microsoft Q&A thread.

The error is:

715-123420

"Our system has detected this request as unusual activity for your account. If you are confident this is in error, please contact support."

In the previous Q&A thread, a Microsoft External Staff moderator collected the required diagnostic information from me via private message and informed me that the case had been routed to the backend team for further investigation.

However, more than one month has now passed without any update, despite multiple follow-up messages.

The issue is still unresolved.

I have already completed the following checks and tests:

  • Microsoft.CognitiveServices: Registered
  • Microsoft.MachineLearningServices: Registered
  • Azure Policy assignments: None
  • Reproduced via both Azure Portal and Azure CLI
  • Reproduced across multiple Azure AI Foundry resources
  • Reproduced across multiple GPT-5.6 models
  • Reproduced in multiple regions, including East US 2 and Japan West
  • The same error occurs regardless of deployment method
  • Another user in the same company and tenant, also using a Visual Studio Enterprise Subscription - MPN, is able to successfully deploy GPT-5.6 models

The affected subscription is also a Visual Studio Enterprise Subscription - MPN.

Because another user in the same tenant with the same subscription type can deploy successfully, this issue appears to be specific to my subscription/account or to its eligibility, trust, or risk validation.

The previous moderator requested and received the following information:

  • Subscription ID
  • Tenant ID
  • Azure AI Foundry Resource Name
  • Full Resource ID
  • Deployment Name
  • Model Name and Version
  • Deployment Type
  • Failed Deployment Timestamp (UTC)
  • Full Azure CLI Error Output
  • Screenshot of the Error
  • Correlation ID
  • Operation ID
  • Client Request ID

The previous moderator confirmed that the case had been routed to the backend team, but I have not received any further status update.

Could another Microsoft Employee or Microsoft External Staff Moderator please review this situation and confirm whether an active backend incident or review case currently exists for my subscription?

If the previous escalation is no longer active, I would appreciate assistance initiating a new backend review.

I can provide the previous Q&A thread URL and all diagnostic information again via private message if required.

Previous Q&A thread:

https://learn.microsoft.com/en-us/answers/questions/5975192/error-715-123420-when-deploying-any-model-in-micro

Thank you very much for your assistance.

Foundry Models
Foundry Models

A catalog of AI models in Microsoft Foundry that you can discover, compare, and deploy using Azure’s built‑in tools for evaluation, fine‑tuning, and inference

0 comments

1 answer

Sort by: Newest
Most helpful Newest Oldest
  1. Rukshan edirisinghe 910 Reputation points
    2026-09-21T03:50:55.6266667+00:00

    Hi @保元 駿

    Community contributor here, not Microsoft staff, so I'll be upfront: nobody outside Microsoft can see whether your backend case is still active, and even moderators here can't view another moderator's private thread. But I can tell you what this error actually is and how to get a trackable escalation, because the one you have has a structural problem.

    First, what you're dealing with: 715-123420 is not a config, quota or RBAC issue. It's a platform-level block by Azure's Real-Time Fraud Protection (RTFP) system, and only the internal RTFP team can review and lift it. Your evidence already proves it's subscription-scoped (colleague in the same tenant, same offer type, deploys fine), and credit-based offers like Visual Studio Enterprise MPN are exactly the profile this system flags most.

    The structural problem: Q&A backend routing gives you no case number, no SLA, and no way to follow up, which is why a month passed in silence. So do this now, in parallel:

    1. Open a support ticket yourself, and here's the part most people miss: fraud/RTFP blocks fall under billing and account management, which is free on every subscription. Portal > Help + support > Create a support request > issue type Billing. Describe it as "subscription incorrectly flagged for unusual activity, error 715-123420 blocking all model deployments, requesting RTFP team review and unblock". Paste the previous Q&A thread URL, all the correlation and operation IDs you already collected, and mention the working colleague subscription as the control case. This gives you a case number you can chase.
    2. Your VS Enterprise subscription also includes technical support incidents (activate at my.visualstudio.com) if you prefer the technical queue.
    3. In this new thread, explicitly ask the moderator for a secure collaboration ticket, that's the mechanism FTE moderators use to open a tracked case on your behalf when the earlier routing has gone stale.
    4. Until it's lifted, stop retrying deployments repeatedly. Extra churn while flagged can extend the hold. And if work is blocked meanwhile, deploy from a pay-as-you-go subscription, those trip RTFP far less than credit-based offers.

    If this helped, please click Accept Answer so others stuck on this error can find it.

    References: https://learn.microsoft.com/en-us/answers/questions/5863907/azure-ai-foundry-model-deployment-fails-with-error https://learn.microsoft.com/en-us/answers/questions/5949303/unable-to-deploy-gpt-image-2-on-azure-ai-foundry-e

    Was this answer helpful?

    0 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.