PRODUCTION SCENARIO
A loan pre-approval model approves applicants from two postal codes at half the rate of other applicants with the same income and credit history.
Which responsible AI principle does this finding most directly violate?
Answering here is anonymous. Nothing is saved unless you sign in.
Show answer and explanation
Answer: Fairness
Fairness requires an AI system to treat everyone fairly and avoid affecting similar groups differently, so people with similar financial circumstances should receive the same recommendation. Approval rates that diverge for applicants with matching income and credit history are a fairness defect.