Security, Networking, Monitoring, and Responsible AI

Key Takeaways

  • Production AI-103 solutions should prefer Microsoft Entra ID, managed identity, and RBAC over embedded API keys.
  • Private endpoints, private link, and managed networking reduce public exposure for Foundry projects, AI Search, Storage, APIs, and other dependencies.
  • Content filters address harmful content categories, while Prompt Shields target direct and indirect prompt-injection attacks.
  • Monitoring and tracing are separate but related: monitoring tracks production health and quality trends, while tracing captures request-level execution evidence.
  • Responsible AI planning includes evaluators, safety evaluations, provenance metadata, approval workflows, tool constraints, and user-feedback loops.
Last updated: June 2026

Security, Networking, Monitoring, and Responsible AI

AI-103 treats security and responsible AI as design requirements, not cleanup tasks. A Foundry app might pass a quick demo with an API key, public endpoints, and no telemetry, but that design will not satisfy production scenarios. The exam favors keyless authentication, scoped authorization, private connectivity, safety filters, observability, and human oversight where actions are sensitive.

Identity and Access

Use Microsoft Entra ID and managed identity when Azure-hosted compute calls Foundry Models, Azure OpenAI, Azure AI Search, Storage, Key Vault, or internal APIs. A managed identity is an Azure-managed service principal, so code can request tokens without storing secrets. Then use role-based access control (RBAC) to grant only the permissions required at the correct scope.

RequirementBetter designRisky design
App calls model deploymentManaged identity with Azure AI or Azure OpenAI data-plane roleAPI key hard-coded in source or app settings with broad access
Project stores reusable service accessProject connection using Entra ID or managed identity where possibleShared admin key pasted into multiple flows
Developer needs project accessAzure AI User or project-scoped roleSubscription Owner for day-to-day development
Secret still requiredStore in Key Vault and retrieve with managed identityStore in repository, prompt flow file, or front-end code

Networking and Isolation

Use private networking when data sensitivity, compliance, or exfiltration risk matters. Private endpoints and private link create private IP paths to services such as Foundry resources, Storage, Azure AI Search, API Management, or private tool endpoints. In stricter designs, disable public network access and make sure every dependency has a private route, not just the model endpoint.

For agents, also constrain tools. Register only the tools the agent needs, authenticate tool calls independently, and require approval before high-impact actions such as sending external messages, updating production records, executing code, or calling privileged APIs.

Safety Controls

Content safety has multiple layers. Basic content filters classify prompts and completions for harm categories such as hate, sexual, violence, and self-harm. Prompt Shields look for direct jailbreak attempts and indirect prompt-injection attacks hidden in retrieved documents, web pages, emails, images, or uploaded files. Protected-material filters can help detect known protected text or code in generated output.

ControlMain purposeExam clue
Content filterBlock or annotate unsafe input and output categoriesHarmful prompt or generated response
Prompt ShieldsDetect jailbreak and indirect prompt injectionRetrieved content tries to override system instructions
Groundedness evaluatorCheck whether an answer is supported by provided sourcesRAG bot invents facts or ignores citations
Approval workflowKeep humans in control of sensitive actionsAgent can spend money, change data, or contact users

Monitoring, Tracing, and Evaluation

Monitoring answers: "Is production healthy?" Foundry monitoring integrates with Azure Monitor and Application Insights to surface token consumption, latency, exceptions, quality scores, safety events, and response trends. Use alerts when cost spikes, latency rises, groundedness drops, or harmful-output metrics breach thresholds.

Tracing answers: "What happened in this one request?" Distributed traces record spans for prompt construction, retrieval, model calls, tool invocations, agent decisions, and downstream services. Traces support debugging, audit review, and provenance because they show the evidence trail behind a response.

Evaluation answers: "Is this system good enough?" Run pre-production evaluation on labeled datasets and use continuous evaluation on sampled production traffic. Include quality metrics, risk and safety evaluators, groundedness, relevance, task completion, and custom business checks. A reliable AI-103 solution combines all three: monitoring for trends, tracing for detail, and evaluation for measurable quality.

Test Your Knowledge

A RAG-powered agent uses managed identity and private endpoints, but users report that retrieved documents sometimes contain hidden instructions telling the agent to ignore its system prompt. Which control most directly addresses this risk?

A
B
C
D