Listen to this Post
Microsoft has announced the general availability (GA) of Threat Protection for AI services in Defender for Cloud, starting May 1, 2025. This new Defender for AI Services plan will enhance security for models in Azure AI and Azure OpenAI Services.
🔗 Reference: Microsoft Defender for AI Services Blog
You Should Know:
Key Features of Defender for AI Services
- Threat detection for AI models in Azure AI & OpenAI
- Anomaly detection in AI workloads
- Integration with Microsoft Defender XDR for unified security
- Compliance & risk assessment for AI deployments
How to Enable Defender for AI Services
1. Access Defender for Cloud:
az security setting list --query "[?name=='MCAS'].enabled"
2. Enable AI Threat Protection:
az security setting update --name "AIProtection" --status On
3. Verify AI Model Protection:
Get-AzSecurityAIStatus -ResourceGroupName "YourRG"
Monitoring AI Security Events
- Check AI-related alerts in Defender:
az security alert list --filter "AssignedTo eq 'AI Security Team'"
- Use KQL queries in Azure Sentinel:
[kql]
SecurityAlert | where ProviderName contains “DefenderAI”
[/kql]
Hardening AI Workloads
- Restrict API access to AI models:
az role assignment create --role "AI Reader" --assignee [email protected]
- Enable MFA for AI administrators:
Set-MsolUser -UserPrincipalName [email protected] -StrongAuthenticationRequirements @{}
What Undercode Say
AI security is critical as adoption grows. Defender for AI Services fills a crucial gap by integrating threat protection directly into Azure AI and OpenAI deployments. Key takeaways:
- Monitor AI models like any other critical asset.
- Use least privilege access for AI APIs.
- Automate threat detection with Defender alerts.
- Audit AI workloads regularly for anomalies.
🔧 Useful Commands:
- List AI security recommendations:
az security recommendation list --query "[?contains(displayName, 'AI')]"
- Check AI service logs:
az monitor activity-log list --resource-type "Microsoft.AI/machineLearning"
- Enable diagnostic settings for AI services:
Set-AzDiagnosticSetting -ResourceId /subscriptions/{sub-id}/resourceGroups/{rg}/providers/Microsoft.AI/machineLearning -Enabled $true
Expected Output:
A secured AI environment with continuous monitoring, automated threat detection, and compliance enforcement for Azure AI and OpenAI services.
🔗 Reference: Microsoft Defender for AI Services Blog
References:
Reported By: Markolauren Ga – Hackers Feeds
Extra Hub: Undercode MoN
Basic Verification: Pass ✅



