Azure AI and LLM Security

This unit focuses on securing large language model (LLM) applications and Azure AI services, including Azure AI Search, across the cloud security lifecycle. Students will learn to apply AI security posture management, integrate LLMs into cloud security architectures, and produce actionable controls and monitoring strategies.

Learning Objectives

• Analyze Azure AI Search capabilities and architecture to identify authentication, data protection, indexing, and query-surface risks specific to enterprise deployments
• Apply the LLM Application Security Lifecycle by mapping security requirements across design, data ingestion, model selection, testing, deployment, and monitoring to create a secure deployment checklist for Azure-based LLMs
• Demonstrate implementation of AI Security Posture Management (ASPM) and Azure-native security controls—such as RBAC, managed identities, network isolation, logging, and alerting—to detect and remediate LLM-specific threats
• Evaluate options for integrating LLM applications into a cloud security architecture and recommend governance, mitigation, and incident-response controls (e.g., prompt/output filtering, provenance, and data handling policies) to reduce organizational risk