Pass Microsoft SC-401 Exam in First Attempt Guaranteed!
Get 100% Latest Exam Questions, Accurate & Verified Answers!
30 Days Free Updates, Instant Download!
Microsoft SC-401 Practice Test 70
Question 690
Legal requires preserving supported Microsoft 365 Copilot interaction records for a defined period under an organizational retention program. Which TWO choices together provide the solution and a valid acceptance test? (Select TWO.)
Explanation
Correct Answer
“Choose the supported Copilot or AI retention policy location and configure appropriate retention settings and scope” would violate the requirement. AI interaction retention is a records-management configuration distinct from DLP blocking of sensitive prompts.
“Sample supported interactions follow the documented retention requirements without assuming chat history equals audit history” would violate the requirement. AI interaction retention is a records-management configuration distinct from DLP blocking of sensitive prompts.
Incorrect Answers
This choice relates to dspm ai usage monitoring, so it cannot substitute for the requested intervention-and-evidence pair. The current implementation is to choose the supported Copilot or AI retention policy location and configure appropriate retention settings and scope. The matched verification must show: Sample supported interactions follow the documented retention requirements without assuming chat history equals audit history. An unrelated control or outcome does not establish the result in this case.
“Create an email sensitivity label and assume it retains all Copilot prompts for seven years” adds a relevant safeguard or verification step rather than weakening the solution. AI interaction retention is a records-management configuration distinct from DLP blocking of sensitive prompts.
This choice relates to dspm for ai initial prerequisites, so it cannot substitute for the requested intervention-and-evidence pair. The current implementation is to choose the supported Copilot or AI retention policy location and configure appropriate retention settings and scope. The matched verification must show: Sample supported interactions follow the documented retention requirements without assuming chat history equals audit history. An unrelated control or outcome does not establish the result in this case.
Question 691
A department labels a design document Confidential but grants all employees unrestricted file permissions; leaders expect Copilot to hide it automatically. Which approach meets the stated business and security constraint?
Explanation
Correct Answer
A visible classification label does not by itself revoke a user's underlying content access; effective permissions and protection matter. Unauthorized test users cannot retrieve the protected source through supported access paths, regardless of AI summarization.
Incorrect Answers
“Correlate available DLP, Audit, activity explorer and supported DSPM for AI events into a case with a documented timeline” does not implement the control that the facts call for. A visible classification label does not by itself revoke a user's underlying content access; effective permissions and protection matter. Instead, review sensitivity label permissions and source access, adding supported protection that actually restricts unintended readers.
“Increase the Confidential watermark opacity and assume it changes Copilot's document authorization” would not meet the requirement. A visible classification label does not by itself revoke a user's underlying content access; effective permissions and protection matter. Instead, review sensitivity label permissions and source access, adding supported protection that actually restricts unintended readers.
“Check and satisfy current DSPM for AI prerequisites, permissions and supported workload integrations” does not implement the control that the facts call for. A visible classification label does not by itself revoke a user's underlying content access; effective permissions and protection matter. Instead, review sensitivity label permissions and source access, adding supported protection that actually restricts unintended readers.
Question 692
A department labels a design document Confidential but grants all employees unrestricted file permissions; leaders expect Copilot to hide it automatically. The current control has not met the requirement. What should the administrator investigate first?
Explanation
Correct Answer
Inspect the label encryption rights, document ACLs and the effective Copilot user's access to the source. A visible classification label does not by itself revoke a user's underlying content access; effective permissions and protection matter. Unauthorized test users cannot retrieve the protected source through supported access paths, regardless of AI summarization.
Incorrect Answers
“Inspect user, file identifiers, destinations, sequence times and gaps between telemetry sources” would not be the most direct first check for this failure. Inspect the label encryption rights, document ACLs and the effective Copilot user's access to the source.
“Inspect connector or browser support, activity times, policy scope and known test submissions” would not be the most direct first check for this failure. Inspect the label encryption rights, document ACLs and the effective Copilot user's access to the source.
“Inspect supported AI workload, retention policy target, user assignments and preserved records” would not be the most direct first check for this failure. Inspect the label encryption rights, document ACLs and the effective Copilot user's access to the source. A visible classification label does not by itself revoke a user's underlying content access; effective permissions and protection matter.
Question 693
A department labels a design document Confidential but grants all employees unrestricted file permissions; leaders expect Copilot to hide it automatically. What observable outcome should the team require before closing the change?
Explanation
Correct Answer
A visible classification label does not by itself revoke a user's underlying content access; effective permissions and protection matter. Unauthorized test users cannot retrieve the protected source through supported access paths, regardless of AI summarization. If this test fails, inspect the label encryption rights, document ACLs and the effective Copilot user's access to the source.
Incorrect Answers
“Sample supported interactions follow the documented retention requirements without assuming chat history equals audit history” alone does not establish all the required behavior. Unauthorized test users cannot retrieve the protected source through supported access paths, regardless of AI summarization. A visible classification label does not by itself revoke a user's underlying content access; effective permissions and protection matter.
“The lead can review approved risk signals while sensitive policy changes require authorized administrators” alone does not establish all the required behavior. Unauthorized test users cannot retrieve the protected source through supported access paths, regardless of AI summarization. A visible classification label does not by itself revoke a user's underlying content access; effective permissions and protection matter.
“A reviewer can distinguish verified activity from unresolved allegations and identify which protections acted” alone does not establish all the required behavior. Unauthorized test users cannot retrieve the protected source through supported access paths, regardless of AI summarization.
Question 694
A department labels a design document Confidential but grants all employees unrestricted file permissions; leaders expect Copilot to hide it automatically. What underlying behavior makes the selected control suitable?
Explanation
Correct Answer
A visible classification label does not by itself revoke a user's underlying content access; effective permissions and protection matter. Review sensitivity label permissions and source access, adding supported protection that actually restricts unintended readers. The decisive evidence is that unauthorized test users cannot retrieve the protected source through supported access paths, regardless of AI summarization.
Incorrect Answers
“AI interaction retention is a records-management configuration distinct from DLP blocking of sensitive prompts” does not fully explain the decisive technical boundary in this case. A visible classification label does not by itself revoke a user's underlying content access; effective permissions and protection matter.
“DSPM for AI permissions should reflect separation between observation and production policy changes” does not fully explain the decisive technical boundary in this case. A visible classification label does not by itself revoke a user's underlying content access; effective permissions and protection matter.
“AI-related exfiltration investigations benefit from correlating product-specific security evidence rather than assuming a single dashboard is complete” does not fully explain the decisive technical boundary in this case. A visible classification label does not by itself revoke a user's underlying content access; effective permissions and protection matter.
Question 695
A department labels a design document Confidential but grants all employees unrestricted file permissions; leaders expect Copilot to hide it automatically. Which TWO choices together provide the solution and a valid acceptance test? (Select TWO.)
Explanation
Correct Answer
“Unauthorized test users cannot retrieve the protected source through supported access paths, regardless of AI summarization” would violate the requirement. A visible classification label does not by itself revoke a user's underlying content access; effective permissions and protection matter.
“Review sensitivity label permissions and source access, adding supported protection that actually restricts unintended readers” would violate the requirement. A visible classification label does not by itself revoke a user's underlying content access; effective permissions and protection matter.
Incorrect Answers
“Increase the Confidential watermark opacity and assume it changes Copilot's document authorization” adds a relevant safeguard or verification step rather than weakening the solution. A visible classification label does not by itself revoke a user's underlying content access; effective permissions and protection matter.
This choice relates to least privilege dspm for ai roles, so it cannot substitute for the requested intervention-and-evidence pair. The current implementation is to review sensitivity label permissions and source access, adding supported protection that actually restricts unintended readers. The matched verification must show: Unauthorized test users cannot retrieve the protected source through supported access paths, regardless of AI summarization. An unrelated control or outcome does not establish the result in this case.
This choice relates to ai service connector and visibility gaps, so it cannot substitute for the requested intervention-and-evidence pair. The current implementation is to review sensitivity label permissions and source access, adding supported protection that actually restricts unintended readers. The matched verification must show: Unauthorized test users cannot retrieve the protected source through supported access paths, regardless of AI summarization. An unrelated control or outcome does not establish the result in this case.
Question 696
An analyst investigates a sensitive file uploaded to an AI site shortly after a bulk SharePoint download by the same user. What should the information security administrator implement?
Explanation
Correct Answer
Correlate available DLP, Audit, activity explorer and supported DSPM for AI events into a case with a documented timeline. Confirm that a reviewer can distinguish verified activity from unresolved allegations and identify which protections acted. If the result fails, inspect user, file identifiers, destinations, sequence times and gaps between telemetry sources.
Incorrect Answers
“Close the case as confirmed data theft without comparing source file activity and AI destination records” does not implement the control that the facts call for. Instead, correlate available DLP, Audit, activity explorer and supported DSPM for AI events into a case with a documented timeline.
“Check and satisfy current DSPM for AI prerequisites, permissions and supported workload integrations” does not implement the control that the facts call for. Instead, correlate available DLP, Audit, activity explorer and supported DSPM for AI events into a case with a documented timeline.
“Assign the appropriate DSPM for AI view and administration permissions separately from broader Purview policy roles” does not implement the control that the facts call for. Instead, correlate available DLP, Audit, activity explorer and supported DSPM for AI events into a case with a documented timeline.
Question 697
An analyst investigates a sensitive file uploaded to an AI site shortly after a bulk SharePoint download by the same user. Before making more changes, which inspection most directly isolates the problem?
Explanation
Correct Answer
Inspect user, file identifiers, destinations, sequence times and gaps between telemetry sources. A reviewer can distinguish verified activity from unresolved allegations and identify which protections acted.
Incorrect Answers
“Inspect supported AI workload, retention policy target, user assignments and preserved records” would not be the most direct first check for this failure. Inspect user, file identifiers, destinations, sequence times and gaps between telemetry sources.
“Inspect the label encryption rights, document ACLs and the effective Copilot user's access to the source” would not be the most direct first check for this failure. Inspect user, file identifiers, destinations, sequence times and gaps between telemetry sources.
“Inspect eligible tenant services, role assignments and baseline data security signals” would not be the most direct first check for this failure. Inspect user, file identifiers, destinations, sequence times and gaps between telemetry sources.
Question 698
An analyst investigates a sensitive file uploaded to an AI site shortly after a bulk SharePoint download by the same user. What test result would confirm that the requirement has been satisfied?
Explanation
Correct Answer
A reviewer can distinguish verified activity from unresolved allegations and identify which protections acted. Correlate available DLP, Audit, activity explorer and supported DSPM for AI events into a case with a documented timeline.
Incorrect Answers
“The approved administrators can access supported DSPM for AI assessment insights and configured controls” alone does not establish all the required behavior. A reviewer can distinguish verified activity from unresolved allegations and identify which protections acted.
“Protected customer-data prompts trigger the intended restriction or alert in the supported AI workflow” alone does not establish all the required behavior. A reviewer can distinguish verified activity from unresolved allegations and identify which protections acted.
“Unauthorized test users cannot retrieve the protected source through supported access paths, regardless of AI summarization” alone does not establish all the required behavior. A reviewer can distinguish verified activity from unresolved allegations and identify which protections acted.
Question 699
An analyst investigates a sensitive file uploaded to an AI site shortly after a bulk SharePoint download by the same user. Which Microsoft Purview principle explains the preferred approach?
Explanation
Correct Answer
An analyst investigates a sensitive file uploaded to an AI site shortly after a bulk SharePoint download by the same user. Inspect user, file identifiers, destinations, sequence times and gaps between telemetry sources. A reviewer can distinguish verified activity from unresolved allegations and identify which protections acted.
Incorrect Answers
“Data loss prevention policies can apply to supported AI prompt interactions when the workload and policy conditions are configured” would not be the most direct first check for this failure. Inspect user, file identifiers, destinations, sequence times and gaps between telemetry sources.
“Data Security Posture Management for AI aggregates applicable AI data security signals and policy workflows; it needs supported prerequisites” would not be the most direct first check for this failure. Inspect user, file identifiers, destinations, sequence times and gaps between telemetry sources.
“A visible classification label does not by itself revoke a user's underlying content access; effective permissions and protection matter” would not be the most direct first check for this failure. Inspect user, file identifiers, destinations, sequence times and gaps between telemetry sources.
Question 700
An analyst investigates a sensitive file uploaded to an AI site shortly after a bulk SharePoint download by the same user. Select TWO: the appropriate protection step and the evidence showing it succeeded.
Explanation
Correct Answer
“A reviewer can distinguish verified activity from unresolved allegations and identify which protections acted” would violate the requirement.
“Correlate available DLP, Audit, activity explorer and supported DSPM for AI events into a case with a documented timeline” would violate the requirement. The appropriate control instead is to correlate available DLP, Audit, activity explorer and supported DSPM for AI events into a case with a documented timeline.
Incorrect Answers
This choice relates to dlp protect sensitive ai prompts, so it cannot substitute for the requested intervention-and-evidence pair. The current implementation is to correlate available DLP, Audit, activity explorer and supported DSPM for AI events into a case with a documented timeline. The matched verification must show: A reviewer can distinguish verified activity from unresolved allegations and identify which protections acted. An unrelated control or outcome does not establish the result in this case.
This choice relates to retention for copilot interactions, so it cannot substitute for the requested intervention-and-evidence pair. The current implementation is to correlate available DLP, Audit, activity explorer and supported DSPM for AI events into a case with a documented timeline. The matched verification must show: A reviewer can distinguish verified activity from unresolved allegations and identify which protections acted. An unrelated control or outcome does not establish the result in this case.
“Close the case as confirmed data theft without comparing source file activity and AI destination records” adds a relevant safeguard or verification step rather than weakening the solution.
Administering Information Security in Microsoft 365 (SC-401) Practice Tests
- Microsoft SC-401 Practice Test 68 - 10 Questions (Last Reviewed 07-Nov-2026)
- Microsoft SC-401 Practice Test 69 - 10 Questions (Last Reviewed 06-Nov-2026)
- Microsoft SC-401 Practice Test 70 - 10 Questions (Last Reviewed 05-Nov-2026)
