80 questions with Azure AI Content Safety tags

Sort by: Updated
Updated Created Answers
1 answer

Unsupported File Type Extension while Following Azure Exercise for RAG-based Solutions

Hey, So I'm following along the exercise: https://learn.microsoft.com/en-us/training/modules/build-copilot-ai-studio/5-exercise I downloaded the provided brochure.zip file, which only contains PDF files. I was able to successfully upload the…

Azure AI Content Safety
Azure AI Content Safety
An Azure service that enables users to identify content that is potentially offensive, risky, or otherwise undesirable. Previously known as Azure Content Moderator.
80 questions
asked 2025-12-05T19:43:18.0966667+00:00
Jonathan Nguyen 0 Reputation points
edited the question 2025-12-06T02:05:27.79+00:00
RAMAMURTHY MAKARAPU 1,125 Reputation points Microsoft External Staff Moderator
0 answers

Internal error when training custom category

I'm trying to use the Content Safety Studio to train the AI on a custom category. I uploaded my .jsonl file to a blob storage and have 64 examples in the following format in the file: {"text": "Message: Temporary Access Token One-Time…

Azure AI Content Safety
Azure AI Content Safety
An Azure service that enables users to identify content that is potentially offensive, risky, or otherwise undesirable. Previously known as Azure Content Moderator.
80 questions
asked 2025-12-05T19:32:19.4066667+00:00
Tony Williams 0 Reputation points
2 answers

azure content safety api not working

I'm getting {"blocklistsMatch":[],"categoriesAnalysis":[]} for all text moderation calls i'm making. It used to work before.…

Azure AI Content Safety
Azure AI Content Safety
An Azure service that enables users to identify content that is potentially offensive, risky, or otherwise undesirable. Previously known as Azure Content Moderator.
80 questions
asked 2025-12-03T22:49:49.3333333+00:00
Susie Park 0 Reputation points
answered 2025-12-05T15:33:11.2133333+00:00
Anshika Varshney 3,795 Reputation points Microsoft External Staff Moderator
1 answer

Azure AI Foundry (Classic) Content Filters not working

Greetings! I tried adding Content Filter to Azure AI Foundry (Classic experience) and it doesn't work. Steps to reproduce: Created an AI Agent with Azure AI Foundry using a pre-created gpt-4.1 model deployment Create a content filter (both…

Azure AI Content Safety
Azure AI Content Safety
An Azure service that enables users to identify content that is potentially offensive, risky, or otherwise undesirable. Previously known as Azure Content Moderator.
80 questions
asked 2025-12-05T10:03:24.7266667+00:00
Nazarii Klymok 0 Reputation points
commented 2025-12-05T13:38:41.9733333+00:00
Anshika Varshney 3,795 Reputation points Microsoft External Staff Moderator
2 answers One of the answers was accepted by the question author.

AI-Driven Cloud Cost Optimization

Hi experts! I want to: Implement a system that uses Azure Cost Management API + ML models (Azure ML) to predict future spend. Auto-trigger rightsizing (scale sets, reserved instances) based on prediction. Deliverable: Self-adjusting cost governance…

Azure AI Content Safety
Azure AI Content Safety
An Azure service that enables users to identify content that is potentially offensive, risky, or otherwise undesirable. Previously known as Azure Content Moderator.
80 questions
asked 2025-09-08T05:48:02.75+00:00
Nidhi Priya 571 Reputation points
commented 2025-12-05T12:07:39.1433333+00:00
Ash Katie 0 Reputation points
0 answers

Enable Zero Data Retention on openai resource

I am trying to enable Zero Data Retention on an openai resource in my org. How should I do this?

Azure AI Content Safety
Azure AI Content Safety
An Azure service that enables users to identify content that is potentially offensive, risky, or otherwise undesirable. Previously known as Azure Content Moderator.
80 questions
asked 2025-12-03T15:26:38.0833333+00:00
Ashwin 0 Reputation points
commented 2025-12-04T09:35:43.4833333+00:00
Anshika Varshney 3,795 Reputation points Microsoft External Staff Moderator
2 answers

Modified Content Filters / Managed Customer access.

[URLs modified to not qualify as a backlink] For our service https://crimeowl(.)ai we use managed Ai instances of various LLM models. Since we are into investigation crime and cold cases we get material that describes crimes as well. We already set…

Azure AI Content Safety
Azure AI Content Safety
An Azure service that enables users to identify content that is potentially offensive, risky, or otherwise undesirable. Previously known as Azure Content Moderator.
80 questions
asked 2025-11-26T11:17:39.9433333+00:00
Martijn van Halen 0 Reputation points
answered 2025-11-26T14:09:12.3966667+00:00
SRILAKSHMI C 10,805 Reputation points Microsoft External Staff Moderator
2 answers One of the answers was accepted by the question author.

Validity of Scan Reports Despite Logged Errors

We ran the baseline scan and attack strategy scan against our model using the Red Team SDK. Although the output was generated, the logs show an error. Should I consider the report valid or invalid?

Azure AI Content Safety
Azure AI Content Safety
An Azure service that enables users to identify content that is potentially offensive, risky, or otherwise undesirable. Previously known as Azure Content Moderator.
80 questions
asked 2025-11-23T02:32:03.5133333+00:00
Karthickumar Karuppiah 80 Reputation points
accepted 2025-11-24T10:21:02.8633333+00:00
Karthickumar Karuppiah 80 Reputation points
1 answer One of the answers was accepted by the question author.

RedTeam SDK: ProtectedMaterial, CodeVulnerability, and UngroundedAttributes Testing

In the RedTeam SDK, I can see that the following categories have been added: ProtectedMaterial 200 CodeVulnerability 389 UngroundedAttributes 200 How can I test these against my model? What does ProtectedMaterial 200, CodeVulnerability 389, and…

Azure AI Content Safety
Azure AI Content Safety
An Azure service that enables users to identify content that is potentially offensive, risky, or otherwise undesirable. Previously known as Azure Content Moderator.
80 questions
asked 2025-11-23T03:15:27.74+00:00
Karthickumar Karuppiah 80 Reputation points
accepted 2025-11-24T09:07:06.36+00:00
Karthickumar Karuppiah 80 Reputation points
1 answer One of the answers was accepted by the question author.

Handling or Disabling Content Filters for Medical Use Case in Agents Service

Hey Everyone, I am using Azure AI Agent Service with OpenAI gpt-4.1 model, and I have a medical use case that requires the agent to handle detecting self harm responses and act appropriately by calling our custom tools. Approaches Used: Configuring…

Azure AI Content Safety
Azure AI Content Safety
An Azure service that enables users to identify content that is potentially offensive, risky, or otherwise undesirable. Previously known as Azure Content Moderator.
80 questions
asked 2025-11-14T18:42:57.0666667+00:00
Omar Elhanafy 20 Reputation points
commented 2025-11-19T16:57:10.87+00:00
Sridhar M 2,675 Reputation points Microsoft External Staff Moderator
2 answers One of the answers was accepted by the question author.

With the Red Team SDK, can we test only safety risks, or can we also test the OWASP Top 10 for LLMs?

The Red Team Agent SDK is working fine for testing risk categories and attack strategies. I want to know whether it is possible to test prompt injection, indirect prompt injection, and data poisoning attacks using the Red Team SDK

Azure AI Content Safety
Azure AI Content Safety
An Azure service that enables users to identify content that is potentially offensive, risky, or otherwise undesirable. Previously known as Azure Content Moderator.
80 questions
asked 2025-11-14T09:40:18.9966667+00:00
Karthickumar Karuppiah 80 Reputation points
edited an answer 2025-11-17T05:32:35.42+00:00
SRILAKSHMI C 10,805 Reputation points Microsoft External Staff Moderator
3 answers One of the answers was accepted by the question author.

AI RED TEAM AGENT unable to get output

Prerequisites: Red Team SDK and other libraries should be installed, and environment variables for CLIENT_ID, CLIENT_SECRET, TENANT_ID, and AZURE_AI_PROJECT should be set. When I run the source code, I don’t get any output.

Azure AI Content Safety
Azure AI Content Safety
An Azure service that enables users to identify content that is potentially offensive, risky, or otherwise undesirable. Previously known as Azure Content Moderator.
80 questions
asked 2025-11-07T07:52:14.2633333+00:00
Karthickumar Karuppiah 80 Reputation points
accepted 2025-11-11T06:25:33.0033333+00:00
Karthickumar Karuppiah 80 Reputation points
1 answer One of the answers was accepted by the question author.

Getting 404 error when trying Azure AI foundry AI Red Teaming

I am trying to work on AI red teaming in AI foundry, and retrieved a sample from documentation even for that even though complete setup is there like AZURE_SUBSCRIPTION_ID, AZURE_RESOURCE_GROUP, AZURE_PROJECT_NAME. here is my code: and error import…

Azure AI Content Safety
Azure AI Content Safety
An Azure service that enables users to identify content that is potentially offensive, risky, or otherwise undesirable. Previously known as Azure Content Moderator.
80 questions
asked 2025-07-30T05:47:28.2666667+00:00
Abhishek Ramdhan Handibag 25 Reputation points
commented 2025-11-07T01:52:52.7166667+00:00
Aryan Parashar 3,380 Reputation points Microsoft External Staff Moderator
1 answer

Groundedness detection through Content Safety is not working

Hi, Content safety's groundedness detection is not working, when I run my code I receive the following output constantly { "ungroundedDetected": false, "ungroundedPercentage": 0, "ungroundedDetails": [] } Also when…

Azure AI Content Safety
Azure AI Content Safety
An Azure service that enables users to identify content that is potentially offensive, risky, or otherwise undesirable. Previously known as Azure Content Moderator.
80 questions
asked 2025-09-20T13:15:29.5566667+00:00
Gebru, J (Jonathan) 0 Reputation points
answered 2025-10-31T15:58:35.8933333+00:00
Manas Mohanty 13,340 Reputation points Moderator
1 answer

Got an email saying ImageGen resource is throttled but couldn't get more details on the problem

Hi I just received an email saying my image generation resource is throttled, for the Model Name ImageGen. I don't know which model is actually causing the issue or what issue is it causing. I couldn't find my details on the issue other than the email I…

Azure AI Content Safety
Azure AI Content Safety
An Azure service that enables users to identify content that is potentially offensive, risky, or otherwise undesirable. Previously known as Azure Content Moderator.
80 questions
asked 2025-10-27T21:10:06.0766667+00:00
John Zhang 0 Reputation points
answered 2025-10-28T03:09:01.11+00:00
SRILAKSHMI C 10,805 Reputation points Microsoft External Staff Moderator
1 answer One of the answers was accepted by the question author.

turn off content filtering

How do i turn off content filtering for azure open AI and cognitive services? My app relies on those programs to assess messages some of which contain harmful or violent messaging.

Azure AI Content Safety
Azure AI Content Safety
An Azure service that enables users to identify content that is potentially offensive, risky, or otherwise undesirable. Previously known as Azure Content Moderator.
80 questions
asked 2025-10-21T17:13:53.46+00:00
Michael Varga 25 Reputation points
edited a comment 2025-10-27T17:42:07.9666667+00:00
Anshika Varshney 3,795 Reputation points Microsoft External Staff Moderator
1 answer One of the answers was accepted by the question author.

Model deployment succeeded, but endpoint responses are inconsistent across regions

I managed to resolve the earlier “ModelNotFound” issue when deploying my custom text classification model in Azure AI Foundry — turned out it was a region mismatch between the training and endpoint resources. Now I’m facing a new problem: while the model…

Azure AI Content Safety
Azure AI Content Safety
An Azure service that enables users to identify content that is potentially offensive, risky, or otherwise undesirable. Previously known as Azure Content Moderator.
80 questions
asked 2025-10-25T20:27:54.9333333+00:00
zhuzin zhuzin 40 Reputation points
accepted 2025-10-25T21:31:22.6066667+00:00
zhuzin zhuzin 40 Reputation points
1 answer

Create new resource in order to use Azure AI labs

I try to creat new resource but I cannot because the details of resource are not available to my subsicribtion, what should I do in this case? By the way, I am instructor and I want to try Azure AI lab but it does not work because that I cannot create…

Azure AI Content Safety
Azure AI Content Safety
An Azure service that enables users to identify content that is potentially offensive, risky, or otherwise undesirable. Previously known as Azure Content Moderator.
80 questions
asked 2025-10-23T09:05:13.7233333+00:00
Ali Alsubaya 0 Reputation points
answered 2025-10-23T13:51:58.46+00:00
Vinodh247 40,031 Reputation points MVP Volunteer Moderator
2 answers One of the answers was accepted by the question author.

Clarification Needed: Is GPT-OSS Model deployed as serverless in Azure cloud Inference Supported in Azure UAE North Region?

Clarification Needed: Is GPT-OSS Model deployed as serverless in Azure cloud Inference Supported in Azure UAE North Region? Is it inference within the UAE, ensuring that prompt and response data do not leave the country? We're exploring compliant…

Azure AI Content Safety
Azure AI Content Safety
An Azure service that enables users to identify content that is potentially offensive, risky, or otherwise undesirable. Previously known as Azure Content Moderator.
80 questions
asked 2025-10-15T11:10:35.5033333+00:00
Jagdish Galipelli 20 Reputation points
accepted 2025-10-16T11:33:50.28+00:00
Jagdish Galipelli 20 Reputation points
1 answer

Tracking ID Q_3X-BQ0 and Incident ID 690203503

Dear Support, We've received an OpenAI violation, which has given us a 24 hr throttle (See below). We want to understand why we have triggered this and how to proceed. Do we need to swap model? We'd like to know the prompt that violated the usage policy.…

Azure AI Content Safety
Azure AI Content Safety
An Azure service that enables users to identify content that is potentially offensive, risky, or otherwise undesirable. Previously known as Azure Content Moderator.
80 questions
asked 2025-09-26T09:05:45.93+00:00
Alex Powell 0 Reputation points
edited an answer 2025-10-08T12:09:25.8633333+00:00
Manas Mohanty 13,340 Reputation points Moderator