This browser is no longer supported.
Upgrade to Microsoft Edge to take advantage of the latest features, security updates, and technical support.
![](data:image/svg+xml, %3Csvg xmlns='http://www.w3.org/2000/svg' height='64' class='font-weight-bold' style='font: 600 30.11764705882353px "SegoeUI", Arial' width='64'%3E%3Ccircle fill='hsl(160, 70%, 25%)' cx='32' cy='32' r='32' /%3E%3Ctext x='50%25' y='55%25' dominant-baseline='middle' text-anchor='middle' fill='%23FFF' %3EDG%3C/text%3E%3C/svg%3E)
Our system has detected this request as unusual activity for your account. If you are confident this is in error, please contact support.
Properties
Stack trace
BatchARMResponseError at ql (https://oai.azure.com/assets/manualChunk_common_core-85a74f98.js:5:283166) at async Do (https://oai.azure.com/assets/manualChunk_common_core-85a74f98.js:5:292341) at async Object.mutationFn (https://oai.azure.com/assets/manualChunk_common_core-85a74f98.js:5:297026)
Target
can someone help here ?
An Azure service that provides access to OpenAI’s GPT-3 models with enterprise capabilities.
![](data:image/svg+xml, %3Csvg xmlns='http://www.w3.org/2000/svg' height='64' class='font-weight-bold' style='font: 600 30.11764705882353px "SegoeUI", Arial' width='64'%3E%3Ccircle fill='hsl(28.799999999999997, 15%, 12%)' cx='32' cy='32' r='32' /%3E%3Ctext x='50%25' y='55%25' dominant-baseline='middle' text-anchor='middle' fill='%23FFF' %3EAV%3C/text%3E%3C/svg%3E)
Hello @Divya Gilla
When this error occurs (Error Code 715-123420 or 555420), it indicates that Azure AI Foundry has blocked GPT model deployment due to a fraud signal, and the request requires review by the appropriate internal team.
To proceed with further investigation and escalation, could you please share the details requested in private message for the same.
Thankyou!
Hi,
can anyone from microsoft help me with this ?
Hi , Still having this issue even after purchasing Microsoft developer support i'm notable to create a support request to fix this issue. can anyone help me ?
Hello @Divya Gilla
Thanks for your details.
We’ve raised INC for this issue, and our team is actively working on it. Please remain patient we’ll provide an update as soon as we receive further information.
Thankyou!
Hello @Divya Gilla
Could you please confirm whether it's alright to close this thread? if yes, please take a minute to accept my answer this helps increase visibility of this question for other members of the Microsoft Q&A community.
Thank you for helping to improve Microsoft Q&A!
Hello @Divya Gilla
We have received an update from the PG team that the issue has been rectified. Analysis and adjustments completed, please try again.
Could you please check from your end and confirm whether the issue is now resolved?
Looking forward to your response.
Thankyou!
Hello Divya Gilla,
Greetings! Thanks for raising this question in the Q&A forum.
The stack trace you shared (BatchARMResponseError originating from oai.azure.com) combined with error code 715-123420 is a well documented and currently recurring issue across many Azure AI Foundry customers, not a misconfiguration on your side. This is not a quota, region, RBAC, or content-policy problem. It happens because Azure's backend Real-Time Fraud Protection (RTFP) system has placed a subscription-level hold on deployment requests. Because the block is applied before provisioning even starts, it explains why the failure is immediate and consistent across models and regions, and why nothing you change in the portal (resource group, region, SKU) will resolve it on its own.
Here is how to get this unblocked.
Confirm this is the platform-level block, not something else Since the error appears the instant you try to deploy or update any model, and the message text matches "715-123420: An error occurred. Please reach out to support for additional assistance," this strongly points to RTFP rather than a per-resource issue. You do not need to recreate resource groups, switch regions, or change SKUs, since none of those actions clear the hold.
Open an Azure Support request (Technical issue type) In the Azure Portal, go to Help + support > Create a support request.
- Explicitly mention error code **715-123420**
- State that this matches the Real-Time Fraud Protection (RTFP) pattern reported in other Microsoft Q&A threads
- Include your Subscription ID, the affected resource name(s), region(s), and the exact model/deployment you attempted (in your case the deployment that produced the BatchARMResponseError)
- Include a Correlation ID or Trace ID if the Foundry portal surfaced one alongside the error, since this speeds up the backend lookup
**Ask for escalation to the Fraud Protection & Security review team** In the description, request that the ticket be routed for manual review by the internal Fraud Protection & Security team, since this specific error code requires that team to clear the subscription-level hold. Front-line support cannot lift this themselves; it needs backend confirmation that your usage is legitimate.
**What to expect afterward** Once the backend team completes review, they typically confirm the tenant or subscription has been added to an allow list and that deployments can proceed normally. This can take some time depending on ticket volume, so keep the support request open and respond promptly if the support engineer asks for additional non-sensitive identifying details.
If this answer helps you kindly accept the answer which will help others who have similar questions.
Best Regards,
Jerald Felix.
Hi Jerald,
Thank you — the RTFP explanation matches our experience precisely: immediate, consistent 715-123420 on deployment create, across models (gpt-5-mini, gpt-5.4-nano, model-router), across portal and CLI, unaffected by region/SKU/resource-group changes.
Current status, still blocked, with one odd data point that may help others: a single gpt-5-mini (Global Standard) deployment unexpectedly succeeded through the new Foundry portal "serverless deployment" flow, and it works — inference calls return normally. But every deployment attempt before and after it, portal or CLI, still fails with 715-123420, confirmed again minutes ago. So the subscription-level hold is clearly still in place; one request appears to have slipped through a different code path rather than the hold being lifted.
On the support path you describe, two practical walls worth documenting:
We have also emailed csgate@microsoft.com (the address Microsoft's own troubleshooting article gives for OpenAI access/gating issues) with the error code, subscription ID, and portal Service Request ID / Trace ID.
Question back to you: when the portal ticket goes through, is naming "Real-Time Fraud Protection" and requesting routing to the Fraud Protection & Security review team in the first message genuinely enough, or is there additional wording that avoids front-line loops? Given front-line cannot lift these, anything that shortcuts triage would help everyone on this error.
Best,
Divya