Control false negatives: Which evidence should receive priority before approval?
Prioritize measuring and limiting false negatives because missed eligible applicants carry greater harm.
The question
A housing administrator uses a screening model to flag applications for manual review. Missing a potentially eligible applicant is more harmful than reviewing an extra eligible applicant. Which evidence should receive priority before approval?
Preparing for AIGP? Take the free 5-min readiness quiz →
- Compare error rates across applicant subgroups before setting a thresholdSubgroup analysis is important for fairness assessment, but it does not by itself prioritize controlling the stated false-negative harm.
- Maximize precision among flagged applicationsHigher precision may reduce unnecessary reviews but can increase missed eligible applicants.
- Optimize overall classification accuracyOverall accuracy treats errors in aggregate and may not control the more harmful missed-eligibility cases.
- Control false negatives ✓False negatives represent missed eligible applicants, the error type identified as more harmful.
The trap
Translate the stated harm into the corresponding error type before selecting performance evidence. How to remember it
Prioritize measuring and limiting false negatives because missed eligible applicants carry greater harm.
How many of these would you get right?
One of 1581 AIGP questions on Certsqill. Take a free five-minute check and see your score per domain — not one number, but which section to open tonight.
Test your AIGP readiness — freeMore Understanding How to Govern AI Development questions
- Use subgroup acceptance criteria and investigate: What evidence should determine approval? →
- Conduct security testing alongside targeted robustness: Which plan is most defensible? →
- Test re-identification risk and real-image subgroup: Before approval, which evidence is most important? →
- All 426 Understanding How to Govern AI Development questions →
Part of the Certsqill AIGP question bank · Understanding How to Govern AI Development ·
Every answer, right and wrong, comes with its own explanation.
