Error 715-123420 – GPT-4.1 Deployment Blocked in Microsoft Foundry

Error 715-123420 requires a Microsoft service-side review; with Basic support, open a free Subscription management request and ask that it be routed to the Foundry/Azure OpenAI security team for review of the “unusual activity” deployment block.

Exact-match Microsoft Q&A cases were resolved only after the product or security team reviewed and unblocked the affected subscription.^1^ ^2^

Open the request without purchasing technical support

Since the Technical request path already stopped at the support-plan requirement:

1. In the Azure portal, select ? (Help) in the global header.
2. Describe the issue as subscription blocked from deploying Microsoft Foundry models.
3. Select Create a support request.
4. Choose the affected subscription and select Subscription management rather than Technical.
5. If offered a closer category, select an account/subscription access or subscription-management category.
6. In the description, explicitly request:
   • Review of Foundry/Azure OpenAI deployment error 715-123420.
   • Escalation to the appropriate security/risk team.
   • Removal of the deployment restriction if it was incorrectly applied.

Azure provides unlimited subscription-management support; paid support is required for ordinary technical cases. Your subscription-level Owner role is also sufficient to create a request.^3^

Suggested ticket text:

Microsoft Foundry rejects deployment of gpt-4.1 version 2025-04-14 with error 715-123420: Our system has detected this request as unusual activity for your account.

The affected Azure Plan subscription is active, and the deployment was attempted in North Central US. Other Foundry model deployment activity succeeds under the subscription. Please route this request for backend security/risk review and remove any incorrectly applied deployment restriction. Failure occurred on October 29, 2026. Correlation/request IDs and resource details are available privately.

Include:

• Subscription and tenant IDs
• Foundry project and AI Services/OpenAI resource names
• Region and deployment type
• Model and version
• Failure time in UTC
• Correlation, operation, request, and trace IDs
• Screenshot or exported Activity Log event
• Confirmation that the subscription is active and that you hold Owner access

Do not put subscription IDs, tenant IDs, or request diagnostics in the public Q&A post. Supply them only through the private support case or a moderator-enabled private-message channel.

Q&A moderator escalation

A Microsoft Q&A moderator may request private details and facilitate escalation, as happened in similar cases, but a public Q&A post is not itself an Azure support case. One exact-code case records a collaboration with the security team and remediation of a temporary backend block.^2^ Another records the product group completing adjustments and asking the customer to retry after one to two hours.^1^

Keep this Q&A thread open and add only non-sensitive information:

• Azure support request number
• UTC failure timestamp
• Whether every GPT-4.1 deployment fails
• Whether the error is also present under Azure portal → Monitor → Activity log

Ask the moderator to associate the thread with the support request and escalate it under the process for 715-123420.

Until the review completes

Avoid repeated deployment attempts, changing regions solely to bypass the restriction, or deleting and recreating resources. Similar cases report that the restriction was removed only by backend review, not by changing deployment configuration.^4^

The successful DeepSeek deployment does not rule out a restriction specific to Azure OpenAI/Foundry model provisioning. GPT-4.1 version 2025-04-14 is a documented Foundry model, while ordinary availability, quota, and deployment-type failures have separate troubleshooting outcomes.^5^ ^6^

After support confirms release of the restriction, wait for the propagation interval they specify, submit one deployment attempt, and verify that the deployment reaches Succeeded.

References

1. Error 715-123420 when deploying model in Foundry - Microsoft Q&A
2. 715-123420: An error occurred. Please reach out to support for additional assistance. - Microsoft Q&A
3. Create an Azure support request
4. Model Deployment error: 715-123420 - Microsoft Q&A
5. Foundry Models sold by Azure (azure-openai)
6. Deployment types for Microsoft Foundry Models