Fairness
Domain: Safety & Ethics
🎯 Motivation
Ensuring AI systems treat all groups equitably, without discrimination based on race, gender, age, or other protected attributes.
📋 Concrete Example
Fairness check: ‘Does the loan AI approve at similar rates for all demographics?’ If approval rate is 70% for men but 50% for women, there is a fairness problem.
🔗 Analogy
A teacher grading exams blindly - they do not know whose paper it is, so they grade fairly. Fairness in AI means the system does not discriminate.