Azure Content Moderator Azure AI Content Safety
AI guardrail and content moderation service that detects harmful, offensive, ungrounded, or prompt-injection content across text, images, and multimodal generative AI workloads.
Naming timeline
- Azure Content Moderator
Legacy text, image, and human-in-the-loop review moderation API under Azure Cognitive Services.
- Azure AI Content SafetyCurrent name
Next-generation AI guardrail and content moderation service replacing Azure Content Moderator.
What replaced Azure Content Moderator?
Azure Content Moderator has been superseded by Azure AI Content Safety, with the latest change in October 2023 (General Availability & Replacement Service). AI guardrail and content moderation service that detects harmful, offensive, ungrounded, or prompt-injection content across text, images, and multimodal generative AI workloads. The underlying legacy API or CLI surface is deprecated, so a technical migration is required.
Also known as
- Content Safety in Foundry Control Plane
- Content Safety Studio
Technical identifiers
Legacy identifiers deprecated. The underlying API or CLI surface for the legacy name is officially deprecated, so this change requires a technical migration rather than a label update.
| ARM Resource Providers (Microsoft.*) | Microsoft.CognitiveServices/accounts (kind: ContentSafety) |
|---|---|
| Azure CLI (az) / PowerShell | az cognitiveservices account create --kind ContentSafety |
| Entra ID / Azure RBAC Roles | Cognitive Services User Cognitive Services Contributor |
| Terraform Resources (azurerm_*) | azurerm_cognitive_account |
Legacy ContentModerator ARM resource kind and REST endpoints are deprecated (retiring March 15, 2027) in favor of kind = 'ContentSafety' and the /contentsafety REST API.
Notes & disambiguation
- Azure Content Moderator was officially deprecated in February 2024 and is scheduled for retirement on March 15, 2027 in favor of Azure AI Content Safety.
- Announced in preview at Microsoft Build in May 2023 and generally available in October 2023, Azure AI Content Safety introduces severity-scored multiclass detection, prompt shields, groundedness detection, and Content Safety Studio (also surfaced as Content Safety in Foundry Control Plane).
- Migrating from Content Moderator requires provisioning a new Microsoft.CognitiveServices/accounts resource with kind = 'ContentSafety' and updating applications to the Azure.AI.ContentSafety SDK and REST APIs.
Sources & official documentation
Catalog last updated . Spotted an error? Suggest an edit