Azure OpenAI Security Controls

This unit covers threat modeling for large language model (LLM) applications and the practical security controls used to harden Azure OpenAI deployments, including OWASP Top 10 risks, access control, audit logging, network isolation, encryption at rest, and content filtering. It builds on foundational concepts and prepares learners to implement monitoring, incident response, and deployment-hardening in subsequent units.

Learning Objectives

• Analyze the OWASP Top 10 risks for LLM applications and produce a prioritized risk register with mitigation strategies for a provided Azure OpenAI scenario
• Apply Azure role-based access control, managed identities, and policy assignments to configure least-privilege access for Azure OpenAI resources in a lab deployment
• Demonstrate configuration of audit logging and monitoring by enabling diagnostic settings, forwarding logs to Log Analytics or storage, and creating alert rules for security-relevant events
• Evaluate network isolation, encryption-at-rest, and content filtering controls and produce a secure architecture diagram that justifies selected controls against compliance and threat models