Checking an AI-made data representation against the source data
Name:
Date:
Class:
A
Quick Questions
Shade in the A or B circle for each question.
1. An AI tool makes a bar chart of monthly sales. What is the best first check?
2. The source data total is 1,240 but the AI summary says the total is 1,420. What should you do?
3. Which is a way to recalculate a key value to check an AI-made summary?
4. An AI chart shows Maths marks on the y-axis labelled 'English'. What is the problem?
5. An AI chart's y-axis starts at 80 instead of 0, which makes small differences look huge. What should you check?
B
Match the Pairs
Draw a line connecting each item on the left to its match on the right.
| Left column | Right column | |
|---|---|---|
| 1=SUM(B2:B9) | ARecalculate a mean from the source data to check an AI-made summary value | |
| 2=AVERAGE(C2:C9) | BPrompt v2: Use column B only. Show the total for each name. | |
| 3Check that an AI chart's labels and axis scale match the source data | CA value that differs from the source was ignored instead of being corrected | |
| 4If the AI output does not match the source, correct the prompt and re-run it | DAxis: Maths marks 0-100, Source: Maths marks 0-100 | |
| 5AI total = 640, so the chart is correct (source not checked) | ERecalculate the total from the source data to check the total an AI tool reported | |
| 6Chart bar for Mia = 59, source mark = 95. Looks fine, so publish. | FThe AI total was accepted without recalculating it from the original data |
C
Definitions
verify against source data
D
Fill in the Blanks
Complete each line of code by filling in the missing parts.
1.
=___(B2:B9)
2.
=_______(B2:B9,">=80")
3.
import statistics
print(statistics.____(marks))
4.
=___(D2:D9)
5.
total = sum(marks)
if total __ ai_total:
print("Fix prompt and re-run")
E
Spot the Error
Shade in the letter of the line that contains an error.
Question 1
A
src = [12, 18, 15, 20]
B
top = min(src)
C
ai_top = 20
D
print(top == ai_top)
Question 2
A
src = [40, 55, 35]
B
ai = [40, 55, 53]
C
for s, a in zip(src, ai):
D
if s == a: print('Fix')
Question 3
A
src_labels = ['Mon', 'Tue']
B
ai_labels = ['Mon', 'Tue']
C
if ai_labels != src_labels:
D
print('Labels match')
Question 4
A
src = [100, 90, '120']
B
ai_total = 300
C
if ai_total != sum(src):
D
print('Re-run prompt')
Question 5
A
src = [6, 8, 10]
B
avg = sum(src) * len(src)
C
print('AI says 8:', avg == 8)
Answer Key
Checking an AI-made data representation against the source data
Section A: Quick Questions
Q1
A — Compare the bar values with the original data
Q2
A — Recalculate the total from the source data and find the error
Q3
B — =AVERAGE(B2:B9) on the original data
Q4
A — The label does not match the data, so the prompt needs correcting
Q5
B — The scale of the axis
Section B: Match the Pairs
Q1
=SUM(B2:B9) → Recalculate the total from the source data to check the total an AI tool reported
Q2
=AVERAGE(C2:C9) → Recalculate a mean from the source data to check an AI-made summary value
Q3
Check that an AI chart's labels and axis scale match the source data → Axis: Maths marks 0-100, Source: Maths marks 0-100
Q4
If the AI output does not match the source, correct the prompt and re-run it → Prompt v2: Use column B only. Show the total for each name.
Q5
AI total = 640, so the chart is correct (source not checked) → The AI total was accepted without recalculating it from the original data
Q6
Chart bar for Mia = 59, source mark = 95. Looks fine, so publish. → A value that differs from the source was ignored instead of being corrected
Section C: Definitions
Q1
verify against source data — Check an AI-made chart or summary by comparing its values, labels and totals with the original data. If they differ, correct the prompt and re-run it.Section D: Fill in the Blanks
Q1
=SUM(B2:B9)Q2
=COUNTIF(B2:B9,">=80")Q3
import statistics
print(statistics.mean(marks))Q4
=MAX(D2:D9)Q5
total = sum(marks)
if total != ai_total:
print("Fix prompt and re-run")Section E: Spot the Error
Q1
B — The AI chart claims the highest value, so you must recalculate the maximum from the source data. min(src) finds the lowest value. It should be top = max(src).
Q2
D — The check should flag values where the AI value differs from the source value. Using == flags the matches instead. It should be if s != a.
Q3
C — The message 'Labels match' should only print when the AI labels are the same as the source labels. Using != reverses the check. It should be ==.
Q4
A — The source values must stay as numbers so sum(src) can recalculate the total. The quotes make 120 text, so sum() causes an error. It should be 120 without quotes.
Q5
B — To recalculate the mean from the source data you divide the total by the number of values. Multiplying gives 72, not 8. It should be sum(src) / len(src).