For Individuals
- Learning & Certification
- Specialist Tracks & Tools
Consulting & Enterprise
- Programmes & Solutions
- Diagnostics & Evidence
Insights
- Analysis & Research
  - Executive Reports In-depth whitepapers for decision-makers
  - Academic Research Peer-reviewed publications and frameworks
- Editorial & Dispatches
  - The C-Suite AI Memo Weekly executive strategy newsletter
  - News & Commentary Press, updates, and market notes
Events
About
Contact
LOGIN

Skip to content

Safe AI for Data Scientists: Evidence, Documentation and Testing Regulators Accept

What Changes for Data Science When AI Becomes Regulated

Quiz: What Changes for Data Science When AI Becomes Regulated

Problem Framing: Stating Intended Purpose and Out of Scope Uses

Quiz: Problem Framing: Stating Intended Purpose and Out of Scope Uses

Dataset Documentation, Provenance and Licence Checks

Quiz: Dataset Documentation, Provenance and Licence Checks

Representativeness, Coverage Gaps and Known Data Errors

Quiz: Representativeness, Coverage Gaps and Known Data Errors

Personal Data, Special Category Data and Lawful Basis

Quiz: Personal Data, Special Category Data and Lawful Basis

Splitting Data Honestly and Avoiding Leakage

Quiz: Splitting Data Honestly and Avoiding Leakage

Choosing Metrics That Match the Decision Being Automated

Quiz: Choosing Metrics That Match the Decision Being Automated

Fairness Metrics and Why They Cannot All Be Satisfied at Once

Quiz: Fairness Metrics and Why They Cannot All Be Satisfied at Once

Subgroup Analysis and Reporting Performance Where It Is Weakest

Quiz: Subgroup Analysis and Reporting Performance Where It Is Weakest

1 of 2

Previous Lesson

Quiz: Fairness Metrics and Why They Cannot All Be Satisfied at Once

Safe AI for Data Scientists: Evidence, Documentation and Testing Regulators Accept
Fairness Metrics and Why They Cannot All Be Satisfied at Once
Quiz: Fairness Metrics and Why They Cannot All Be Satisfied at Once

Time limit: 0

Quiz Summary

0 of 5 Questions completed

Questions:

Information

You have already completed the quiz before. Hence you can not start it again.

Quiz is loading…

You must sign in or sign up to start the quiz.

You must first complete the following:

Results

Quiz complete. Results are being recorded.

Results

0 of 5 Questions answered correctly

Your time:

Time has elapsed

You have reached 0 of 0 point(s), (0)

Earned Point(s): 0 of 0, (0)
0 Essay(s) Pending (Possible Point(s): 0)

Categories

Not categorized 0%

Review
Answered
Correct
Incorrect

Question 1 of 5

1. Question

Why cannot all fairness metrics be satisfied at once?

Because fairness is subjective
Correct Incorrect Correct answer

Because of impossibility of equalization: if groups have different base rates, you cannot simultaneously achieve equal accuracy, equal positive rate and equal false positive rate
Correct Incorrect Correct answer

Because models are inherently biased
Correct Incorrect Correct answer

Because fairness only applies to high-risk decisions
Correct Incorrect Correct answer

Correct / 1 Points

Incorrect / 1 Points
Question 2 of 5

2. Question

If a loan model approves 70% of men and 60% of women, which fairness metric is violated?

Equal accuracy
Correct Incorrect Correct answer

Equal false positive rate
Correct Incorrect Correct answer

Equal approval rate (demographic parity)
Correct Incorrect Correct answer

Equal opportunity
Correct Incorrect Correct answer

Correct / 1 Points

Incorrect / 1 Points
Question 3 of 5

3. Question

What does equal false negative rate mean?

The model makes no false negatives
Correct Incorrect Correct answer

The model has the same false negative rate across groups
Correct Incorrect Correct answer

False negatives are worse than false positives
Correct Incorrect Correct answer

The false negative rate equals the false positive rate
Correct Incorrect Correct answer

Correct / 1 Points

Incorrect / 1 Points
Question 4 of 5

4. Question

In a medical diagnostic model, what fairness metric might you prioritise?

Equal overall accuracy
Correct Incorrect Correct answer

Equal false negative rate (equal rate of missing disease across groups)
Correct Incorrect Correct answer

Equal approval rate
Correct Incorrect Correct answer

Equal model speed
Correct Incorrect Correct answer

Correct / 1 Points

Incorrect / 1 Points
Question 5 of 5

5. Question

Should you document your fairness metric choice?

No, fairness is inherent to the model
Correct Incorrect Correct answer

Yes, explain which fairness definition you chose and why it is appropriate for your use case
Correct Incorrect Correct answer

Only if challenged
Correct Incorrect Correct answer

Only for sensitive attributes
Correct Incorrect Correct answer

Correct / 1 Points

Incorrect / 1 Points

Login

Accessing this course requires a login. Please enter your credentials below!

Username or Email Address

Password

Remember Me

Continue with Facebook

Continue with Google

Lost Your Password?

Register

Don't have an account? Register one!

Register an Account

Keep building practical AI skills

Short, practical courses from Tesseract Academy, founded by Dr Stelios Kampakis, CStat.

Explore Individual Learning

© 2026 Tesseract Academy. All Rights Reserved.