The provided data failed validation. contains CAPTCHA
Hello, I'm trying to fine-tune GPT4o with image data, but I'm encountering an issue: most of images in my training file are being flagged as "containing CAPTCHAs". Images looks fine. This is the first time I've seen this problem in training…

Content Safety Studio & Cognitive Services
I working through the AI900 labs and am getting the following error when trying to execute some of the sample Content Safety Studio queries. Your account does not have access to this resource, please contact your resource owner to get access. I have…
How to fix resource access that my own resource??
Error Your account does not have access to this resource, please contact your resource owner to get access.
Azure AI Content Safety - Default Blocklist
Is there a default blocklist that I can use when analyzing text with the Azure AI Content Safety API? I am not looking to create and manage my own blocklist, but hoping to see if Azure has a default built-in one already that can be used to block obvious…
System _service^exception
The provided image shows a text editor interface, likely found within a forum or content management system. It's designed for users to input detailed descriptions of problems or issues they're encountering. Purpose: To allow users to thoroughly…
Content safety - prompt shield does it really work?
I am trying the content safety api to test for prompt attacks, I don't understand why it always returns attacks detected=true no matter how I change 'userprompt'. Can anyone tell me, why the code below returns attacks detected? Testing here gives the…
RAI checks when training a fine-tuned model in Azure AI
Hi, Our organization has approval to modify content filters, which we can apply when deploying a model. However, when first training a fine-tuned model, the job fails because some of the training data triggers RAI checks: status : Validation file:…
Custom content filters and blocklists for GPT 4o in Azure OpenAI
I have GPT 4o deployed in North Central US region in my Azure OpenAI resource. I have created a custom content filter with custom blocklists (using instructions from https://learn.microsoft.com/en-us/azure/ai-services/openai/how-to/use-blocklists) via…
Understanding Content Safety Blocklists & KPI Monitoring in Azure OpenAI Enterprise
Hello Azure Community, I’m currently using Azure OpenAI’s Enterprise product and have some questions regarding content safety, blocklists, and KPI monitoring. Content Safety & Blocklists: I understand that the Enterprise version may have additional…
What is the Difference Between Azure Content Safety Studio and Built-in Content Safety in Azure AI Model Deployment
I'm trying to understand the distinction between Azure Content Safety Studio and the content safety options available when deploying AI models in Azure AI. When deploying a model in Azure AI, I noticed that content safety is an option to filter harmful…
How to identify part of prompt triggering "Azure AI Content Safety"
Reaching to you in order to understand why the following exchange is being filter as "sexual" through the Azure AI Content Safety : Description of the flagged content. User input : "Je souhaite vendre et j'ai envie de le faire…
Groundedness detection does not detect anything
When I call the detect groundedness API in the Content Safety service, it always gives the same following response no matter the text or groundingSources: { "ungroundedDetected": false, "ungroundedPercentage": 0, …
Modified Content Filtering form rejected
I filled the form: “Azure OpenAI Limited Access Review: Modified Content Filtering”. But it was rejected (Application ID: 2879423). We have been having constant problems with Azure OpenAIs content policy filters (they are too strict) and wanted to have…
AI service content safety giving 404 error
Hi All, Have created an AI Service content safety instance that within Content Safety Studio is working as expected. However, when I try and set it up in Copilot studio with a HTTP request using the endpoint and keys etc (following this guide…

Content Safety availability in SEA
Is "Content Safety" available on Azure's Data Centers in Singapore? I don’t see it in the table. Thanks!
Error "Subdomain does not map to a resource".
I am doing "Microsoft Learn Challenge | Ignite Edition: Build trustworthy AI solutions on Microsoft Azure" and in the module "Moderate content and detect harm in Azure AI Foundry by using Content Safety" in the "Exercise-Text…
How to remove content filtering from Azure Open AI Deployment
Hello, I would like to ask for clarification regarding the removal of content filtering in an OpenAI deployment. I currently have two default profiles and have created one with all settings set to "low." However, I would like to completely…
Does azure-ai-contentsafety client library have functionality to call promptShield and protectedMaterialDetection endpoint? If not when will it come?
Does azure-ai-contentsafety client library have functionality to call promptShield and protectedMaterialDetection end? If not when will it come? I want to call Azure AI content safety services' promptShield and protectedMaterial endpoint through Azure…
Does azure-ai-contentsafety client library have functionality to call promptShield and protectedMaterialDetection endpoint? If not when will it come?
Does azure-ai-contentsafety client library have functionality to call promptShield and protectedMaterialDetection endpoint? If not when will it come? I want to call Azure AI content safety's promptShield and protectedMaterialDetection endpoint through…
How to remove trained custom categories in Azure AI Studio
I've been working with custom categories in Azure AI Studio for content safety and have successfully trained some models. However, I've run into a limitation where each Content Safety Resource only allows a certain number of custom categories. Currently,…