Securing Microsoft 365 Copilot Agents with Purview and Defender
Short answer
AD You should include Microsoft Purview for data protection and governance, and Microsoft Defender for identifying and mitigating AI-related security risks.
Question
You need to design a security solution for the agents. The solution must meet the following requirements:
Identify and mitigate potential risks that relate to AI use.
Protect AI apps and the sensitive data processed or generated by the agents.
Support responsible AI governance by retaining and logging interactions, detecting policy violations, and investigating incidents.
Which two components should you include in the design? Each correct answer presents part of the solution.
NOTE: Each correct selection is worth one point.
Key Takeaways
The question tests the distinction between data governance (Purview) and threat/risk defense (Defender) specifically within the context of M365 Copilot agents.
Learn how to design a security solution for Microsoft 365 Copilot agents using Microsoft Purview and Microsoft Defender to meet governance and risk mitigation requirements.
Common mistake: Choosing Azure AI Content Safety (B) is common because it relates to 'responsible AI,' but it only handles content filtering and lacks the broader data protection and incident investigation features required.
Expert Analysis
The scenario involves securing Microsoft 365 Copilot agents, which operate within the Microsoft 365 ecosystem. The requirements focus on three pillars: risk identification/mitigation, data protection, and responsible AI governance (logging, policy violations).
Option A (Microsoft Purview) is essential for the second and third requirements. Purview provides Data Loss Prevention (DLP) to protect sensitive data processed by agents, along with compliance policies, retention labels, and audit logging. It is the primary tool for ensuring responsible AI governance regarding data privacy and regulatory compliance.
Option D (Microsoft Defender) addresses the first requirement. Specifically, Microsoft Defender for Cloud Apps (part of Microsoft Defender XDR) identifies and mitigates risks related to AI use by monitoring agent behavior, detecting anomalies, and investigating security incidents. It provides the threat detection and response capabilities needed to secure the AI apps themselves.
Why not B? Azure AI Content Safety is a specific API/service for filtering harmful content (hate speech, violence). While it contributes to responsible AI, it does not provide the broad governance, data protection, or incident investigation capabilities required by the prompt. Comments noting that "Defender is aligned with Copilot Studio agents" highlight the correct operational alignment for M365 Copilot security.
Why not C? RBAC controls access but does not inherently mitigate AI-specific risks, log interactions for governance, or detect policy violations in the manner described.
Official Reference
This link opens the official vendor documentation in a new tab.
Exam Strategy
When designing solutions for Microsoft 365 Copilot, always map 'data protection' and 'governance/logging' to Microsoft Purview, and 'risk identification' and 'threat detection' to Microsoft Defender. Avoid selecting specialized APIs like Content Safety unless the scenario strictly limits itself to content filtering.
Frequently Asked Questions
Why is Microsoft Defender used instead of just Purview?
Does Azure AI Content Safety cover all responsible AI needs?
Discussion 8 contributions
Compiled from public community threads and reader submissions, lightly edited for clarity. Handles and vote tallies are shown as recorded, and comments reflect the views of individual contributors.
From public community threads 8
-
ppolychron 👍 5A and D for me since we are talking about M365 Copilot Agents. DSPM for AI ,DLP, M365 Audit Logs, Insider risk management, Communication Compliance, all valuable tools in Purview. Synching alerts from Purview to Microsoft Defender (XDR) will let you investigate incident from there.
-
omath 👍 2Correct answers: A. Microsoft Purview and B. Azure AI Content Safety These are the only two components that together satisfy all three requirements: risk identification, sensitive‑data protection, and responsible AI governance for Microsoft 365 Copilot agents. Purview provides the governance, auditing, and data‑protection controls that Copilot agents rely on. Purview is the only Microsoft product that logs Copilot interactions and enforces data‑protection policies across the Microsoft 365 ecosystem. Azure AI Content Safety provides AI‑specific risk detection. This directly satisfies the requirement to identify and mitigate AI‑related risks.
-
Ninashnash 👍 2Requirement1: Identify and mitigate potential risks that relate to AI use ---> Microsoft Defender Requirement2: Protect AI apps and the sensitive data processed or generated by the agents ---> Microsoft Defender (Defender for Cloud Apps/Defender for AI services) Requirement3: Support responsible AI governance by retaining and logging interactions, detecting policy violations, and investigating incidents. ----> Microsoft Purview
-
Jedi 👍 2I think A and D are correct. B is aligned with Foundry. Defender is aligned with Copilot Studio agents https://techcommunity.microsoft.com/blog/microsoftthreatprotectionblog/protect-copilot-studio-ai-agents-in-real-time-with-microsoft-defender/4446560
-
Ninashnash 👍 1Exam Tip: Purview = Governance, Compliance, Audit, Retention Defender = Security, Threat Detection, AI Risk Protection Content Safety = Harmful Content Filtering RBAC = Access Control Only
Show 3 more comments
-
Newb007 👍 1A and B. D is related but more adjacent security not really AI....to bad because I love D ;) "“identify and mitigate potential risks that relate to AI use.” is more B
-
Luweho 👍 1Content Safety is a very narrow solution to identify harmful or adult-only content etc.
-
madimad 👍 1AD is correct answer
Reader comments & corrections
No reader corrections yet — spotted an error or have a note? Share it below.