PL-900 Practice Question: Demonstrate the capabilities of Microsoft Copilot Studio
A company has deployed a custom copilot built with Microsoft Copilot Studio. Users report that the copilot sometimes provides outdated information. The knowledge sources are SharePoint sites and public websites. The copilot uses Generative AI to formulate answers. What should the administrator do to improve the accuracy of the copilot's responses?
⚠ Common exam trap
It's easy for candidates to confuse user-facing features like 'Show citations' with data freshness mechanisms, or assume that disabling generative AI will fix accuracy issues, when in fact the root cause is the static nature of the knowledge snapshot.
Answer choices
Why each option matters
Answer the question above first, then reveal the full breakdown to understand why each option is right or wrong.
Correct answer & explanation
✓
Refresh the knowledge sources periodically
The copilot's knowledge sources (SharePoint sites and public websites) are not automatically refreshed in real time. By periodically refreshing these sources, the administrator ensures the copilot's generative AI has access to the most current content, directly addressing the issue of outdated information.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✗
Enable the 'Show citations' option
Why it's wrong here
Showing citations only displays the source links behind a generated answer; it does not refresh stale SharePoint or website content, so outdated responses persist. Citations suit auditing and user trust when answers are already accurate. The accuracy problem stems from knowledge-source indexing, which requires reindexing or updating the sources instead.
- ✗
Reduce the number of topics
Why it's wrong here
Reducing topics trims conversational triggers, not the stale content the generative answers draw from SharePoint and public websites. It is tempting because topic limits can affect performance, and pruning them is valid when a copilot has too many overlapping triggers or hits orchestration limits — but here the fix is refreshing or re-indexing the knowledge sources.
- ✓
Refresh the knowledge sources periodically
Why this is correct
Refreshing knowledge sources periodically re-indexes the SharePoint sites and public websites, so the generative AI answers draw on current content rather than stale cached data. This directly addresses the outdated-information constraint in the stem, since Microsoft Copilot Studio retrieves grounding data from those connected sources at query time.
- ✗
Turn off Generative AI for responses
Why it's wrong here
Disabling generative AI prevents the copilot from synthesising answers at all, so stale knowledge-source content still surfaces verbatim rather than being resolved. It is tempting because generative mode is what hallucinates or paraphrases, and switching it off suits deterministic, scripted flows with fixed responses — not a knowledge-grounded copilot needing fresher retrieval.
Go deeper
Related to this question
About these practice questions
One of 701 original PL-900 practice questions on Courseiva, each with a full explanation and wrong-answer analysis — not exam dumps or protected exam content. Learn why practice questions differ from exam dumps →
JA
Written by Johnson Ajibi, MSc IT Security
Senior Network & Security Engineer · founder of Courseiva
This PL-900 practice question is part of Courseiva's free Microsoft certification practice question bank. Courseiva provides original exam-style practice questions with explanations, topic-based practice, mock exams, readiness tracking, and study analytics to help learners prepare for the PL-900 exam.