Give the customer a plain-language reason: Which additional control is most direct for the customer-facing
The customer needs an understandable explanation of the individual outcome; the supervisor audit gap is already addressed.
The question
A customer-service chatbot denies a refund and displays a technical feature-importance chart. Supervisors already retain and can use that chart to audit patterns across cases; the affected customers cannot understand it. Assume reliability, security, safety, fairness, documentation, and human review are adequate. Which additional control is most direct for the customer-facing gap?
Preparing for AIGP? Take the free 5-min readiness quiz →
- Add more technical variables to the retained chartMore technical detail may assist specialist analysis, but it would not make the existing explanation understandable to customers.
- Publish monthly refund-denial totalsAggregate totals may support system-level transparency, but cannot explain the particular denial experienced by one customer.
- Route every refund request to manual reviewManual review changes the decision process, but does not directly provide an understandable reason for the chatbot’s current outcome.
- Give the customer a plain-language reason ✓A plain-language reason provides the affected customer with an understandable explanation of the individual refund outcome, while the supervisors’ audit need is already met.
The trap
Identify which audience still lacks usable information; do not select a control for an audience whose need is already met. How to remember it
The customer needs an understandable explanation of the individual outcome; the supervisor audit gap is already addressed.
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 the Foundations of AI Governance questions
- Measure false-negative rates separately for each group: Which additional control most directly investigates →
- Model A is supervised: Which classification follows from these training methods? →
- Supervised learning plus reinforcement learning: What classification is supported? →
- All 337 Understanding the Foundations of AI Governance questions →
Part of the Certsqill AIGP question bank · Understanding the Foundations of AI Governance ·
Every answer, right and wrong, comes with its own explanation.
