Fabrikam Legal wants generative answers in Copilot Studio grounded only on an internal SharePoint knowledge library of approved policies. Users must not receive model answers invented from general training data when no policy chunk is found. Which configuration best meets this requirement?
Select an answer to reveal the explanation.
Short Explanation and Infographic
Think of knowledge-only generative answers like a closed-book exam with an open binder — the agent may only quote from the binder (SharePoint). If the binder has nothing, it says so instead of inventing an answer from memory.
Full explanation below image
Full Explanation
Generative answers in Copilot Studio retrieve relevant chunks from configured knowledge sources and ground the LLM response on those chunks. Enterprises often require knowledge-only behavior so the model does not "hallucinate" policy from general training data.
Option A is correct. Adding the approved SharePoint library and configuring generative answers to answer from knowledge (with a no-content fallback message) keeps responses grounded and controllable.
Option B is wrong because Bing/public web expands the answer surface beyond approved policies and can introduce non-authoritative legal content.
Option C is wrong because a system prompt alone does not enforce retrieval grounding; without knowledge sources, the model can still fabricate plausible policy language.
Option D is wrong because removing generative answers and knowledge defeats the goal of policy Q&A; classic topics alone would require hand-authored Q&A for every policy question.
Exam tip: Grounding = knowledge sources + generative answers settings that prefer (or require) retrieved content, with a safe fallback when nothing matches.