Build in human oversight so a qualified reviewer can: Which design practice addresses this?
Human oversight is the design practice that lets a qualified reviewer monitor and override the system's outputs before they take effect.
The question
While designing a high-stakes medical triage AI, the team is applying policies, procedures, and ethical considerations to the build. They specifically want to ensure a qualified person can catch and override an erroneous recommendation. Which design practice addresses this?
Preparing for AIGP? Take the free 5-min readiness quiz →
- Build in human oversight so a qualified reviewer can monitor, interpret, and override the system's outputs before they take effect on patients. ✓Human oversight is the design practice that lets a competent person monitor and override outputs, exactly the safeguard the team wants.
- Maximize the model's accuracy metric on the validation set, since a sufficiently accurate system removes the practical need for a person to intervene at all.High accuracy reduces but never eliminates errors, and it provides no mechanism for a person to catch and override a wrong recommendation.
- Compress the model so it runs faster at the point of care, since low latency is what ensures clinicians can act on its recommendations without any delay.Faster inference improves responsiveness but does nothing to enable review or override of erroneous outputs.
- Encrypt the patient data used by the system in transit and at rest, since strong data protection is the safeguard that ethical medical AI design demands.Encryption protects confidentiality but is unrelated to enabling a human to intercept and correct a flawed recommendation.
The trap
Believing a highly accurate or fast model removes the need for a human able to override its outputs. How to remember it
Human oversight is the design practice that lets a qualified reviewer monitor and override the system's outputs before they take effect.
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
- Identify and evaluate the system's potential effects: What is the core purpose of that assessment at this →
- A probability/severity harms matrix that plots each risk: Which tool from the objective's examples is designed →
- Record the rationale for material design decisions: Given III.A.5, what documentation practice during design →
- 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.
