Provide an authorized: Which design decision is most defensible?
Serious possible harm and uncertainty require a timely human path with real authority to intervene.
The question
An EU public-benefit agency is designing an AI eligibility recommendation tool. A denial could cause serious hardship, evidence about edge cases is incomplete, and reviewers currently cannot reverse recommendations. Which design decision is most defensible?
Preparing for AIGP? Take the free 5-min readiness quiz →
- Sample decisions for supervisory review after issuance.Post-decision sampling can identify patterns, but it does not give an affected person timely intervention against an erroneous denial.
- Require reviewers to acknowledge every recommendation.Acknowledgment does not provide authority, time, or information to correct a harmful recommendation and can encourage rubber-stamping.
- Postpone deployment until every edge case is known.Further testing may be appropriate, but complete knowledge is unlikely; it does not replace designing a workable intervention and appeal mechanism.
- Provide an authorized, timely human appeal and override path. ✓Given serious possible hardship and incomplete evidence, competent reviewers need information, time, and authority to intervene before or after a disputed denial.
The trap
A reviewer is not effective oversight unless the reviewer can act before or after harm occurs. How to remember it
Serious possible harm and uncertainty require a timely human path with real authority to intervene.
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
- Claimant outcomes and distribution of delays: Which outcome should governance examine first? →
- Revise requirements using the driver feedback: What should the team do before continuing the pilot? →
- Select the least resource-intensive model meeting safety: Which decision best reflects the tradeoff? →
- 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.