Identifying Overreliance in Microsoft 365 Copilot Usage
Short answer
B The scenario exemplifies overreliance because the merchandiser acts on Copilot's suggestions without verifying results or applying human judgment.
Question
You prompt Microsoft 365 Copilot to suggest which products to stock based on historical sales data. Without reviewing the suggestions or checking current market trends, you place a large order based solely on the output of Copilot.
What is this an example of?
Key Takeaways
The main trap is confusing overreliance with fabrication; focus on the user's failure to verify rather than the AI's potential error.
This question tests recognition of overreliance risks in Microsoft 365 Copilot scenarios where users accept AI outputs without verification.
Common mistake: Choosing fabrication (C) because one assumes the AI must have generated wrong data, but the question highlights the user's blind acceptance.
Expert Analysis
Identifying the Core Risk
The scenario describes a merchandiser who accepts Microsoft 365 Copilot's product suggestions without verification or consideration of external factors like current market trends. This behavior is a textbook example of overreliance (Option B). In the context of responsible AI and Microsoft 365 Copilot, overreliance occurs when users trust the AI's output too blindly, bypassing necessary human judgment and validation steps.
Why Other Options Are Incorrect
- Fabrication (Option C) refers to the AI generating false or inaccurate information (hallucinations). While the suggestions might be fabricated, the question focuses on the user's action of accepting them without review, not the nature of the data itself.
- Prompt injection (Option D) involves malicious inputs designed to manipulate the AI's behavior, which is not described here.
- Verification (Option A) is the act of checking the output, which the user explicitly failed to do.
Operational Consequence
Failing to recognize overreliance can lead to significant business errors, such as ordering incorrect inventory. The exam tests your ability to identify this risk to emphasize the importance of human-in-the-loop validation.
Adjacent Decision
If the prompt had included instructions to ignore previous safety guidelines, it would be prompt injection. If the AI had provided specific, wrong sales numbers that didn't exist, it might be fabrication. Here, the key is the user's lack of due diligence.
Exam Strategy
When analyzing Copilot scenarios, always look for the user's level of engagement. If they accept output without checks, it's overreliance. If they input malicious text, it's prompt injection. If the AI lies, it's fabrication. Prioritize the human action described.
Frequently Asked Questions
What is the difference between overreliance and fabrication in Copilot?
How does Microsoft 365 Copilot prevent overreliance?
Community Votes
100% of anonymous learners picked answer B. Votes are pick records left by other test-takers — they are not the verified answer.
Discussion 2 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 2
-
Scooter47 👍 4 Answered BThis scenario is an example of overreliance on AI, often specifically referred to as automation bias.
-
TheMCT 👍 1 Answered BThis is an example of overreliance because the merchandiser accepts and acts on Microsoft 365 Copilot's recommendations without applying human judgment, verifying the results, or considering additional information such as current market trends.
Reader comments & corrections
No reader corrections yet — spotted an error or have a note? Share it below.