Can you verify the AI-assisted work sample?
Check the written rule, test its boundary and make the result repeatable.
The question
A fictional finance operations work sample asks for a flag table and a short check note. The assignment expressly permits AI help with the formula. You own the submitted rule and results.
Flag a positive discrepancy of $100 or more. Keep each signed amount in the table.
| Difference ($) | |
|---|---|
| Row A | 92 |
| Row B | 100 |
| Row C | 104 |
These are synthetic examples.
Flag when difference > 100.
“Can you verify the AI-assisted work sample?”
Not saved by this site. Your browser may restore text when you navigate back.
A strong answer, step by step
Open each step in turn. This is one strong answer, not the only one.
Step 1: Which row exposes the error?
What would you do here? Open the move you’d make.
Row C at $104.
Tempting, but it misses something. Both the written rule and the draft flag this row, so it cannot expose the boundary error.
Row B at $100.
The strong move. The brief includes equality and the draft excludes it.
Row A at $92.
Tempting, but it misses something. Both rules leave this row unflagged.
See how a strong candidate says it
“Row B is exactly $100. It meets the brief’s threshold and fails the AI draft’s test.”
Why it works: Checking the boundary distinguishes greater than from greater than or equal to.
Step 2: What should the first table show?
What would you do here? Open the move you’d make.
Flag rows B and C, and show the stated threshold.
The strong move. Both amounts meet the positive threshold in the brief.
Flag only row C because the AI formula runs.
Tempting, but it misses something. A runnable formula can still misstate the requirement.
Delete the $92 row before the reviewer sees it.
Tempting, but it misses something. That removes a useful example below the threshold.
See how a strong candidate says it
“Flag +$100 and +$104 using difference ≥ 100. Keep +$92 in the table without a flag.”
Why it works: The table applies the inclusive rule and lets the reviewer see all the inputs.
Step 3: Which rows does the revised rule flag?
The reviewer adds a −$120 credit. The revised rule flags discrepancies of $100 or more in either direction.
What would you do here? Open the move you’d make.
Keep the positive rule and flag the same two rows.
Tempting, but it misses something. The credit now meets the magnitude threshold.
Compare absolute amounts and flag B, C and the credit.
The strong move. The magnitude check covers the updated rule while the table preserves direction.
Flag every credit, regardless of amount.
Tempting, but it misses something. The revised rule uses the same $100 threshold in both directions.
See how a strong candidate says it
“Compare absolute differences with the inclusive $100 threshold. Flag +$100, +$104 and −$120 and retain their signs.”
Why it works: The reviewer expanded the rule to credits. A positive-only check misses the new credit.
Step 4: What makes the result reviewable?
What would you do here? Open the move you’d make.
Submit a screenshot of the assistant’s response.
Tempting, but it misses something. A screenshot leaves the final rule and tests unclear.
Submit only the three flags without their inputs.
Tempting, but it misses something. The reviewer cannot tell which amounts passed the threshold.
Give the signed rows, the inclusive rule and your test note.
The strong move. The artifact is usable by someone reviewing the work.
See how a strong candidate says it
“Submit the rule, signed inputs and flags. Include the $100 boundary and the credit test. State what AI suggested and what you changed after testing.”
Why it works: The reviewer can reproduce the decisions and see which part of the work you checked.
The answer, in short
- Decision
- Use the inclusive magnitude threshold and preserve the signed differences.
- Because
- The revised rule includes both debits and credits at $100 or more.
- What would change it
- A different threshold or a separate rule for credits.
Common traps
- Checking only examples above the boundary.
- Reporting the AI output without testing it.
Check your answer
- What did the brief permit?
- Which row tests equality?
- How does the credit change the rule?
- Can someone repeat the check?
Last reviewed Nov 10, 2026