The model needs context-specific subgroup testing | AIGP
7-day money-back guarantee — full refund within 7 days of purchase if you've completed under 20% of the questions. See pricing →
Certifications Tools Flashcards Career Paths Exam Guides Blog Pricing For Teams About

Language

✓ EnglishDeutschEspañolFrançaisPortuguês
Check readiness — free →

The model needs context-specific subgroup testing: What does the evidence most directly support?

AIGP Understanding How to Govern AI Development Medium

The model card supports limited validation claims; unequal pilot outcomes require contextual subgroup testing and continuing monitoring.

The question

A housing administrator reviews a model card reporting strong validation accuracy and a separate pilot showing unequal error rates for applicants from one neighborhood. The intended use is application triage, and human review remains available. What does the evidence most directly support?

Preparing for AIGP? Take the free 5-min readiness quiz →

  1. The neighborhood result proves unlawful discrimination.
    Unequal error rates warrant investigation, but these observations alone do not establish a legal conclusion or identify its cause.
  2. Human review removes the need for additional model evaluation.
    Human oversight can mitigate harms, but nominal review does not establish competent intervention or validate model outcomes.
  3. The model card establishes acceptable deployment performance.
    Model-card accuracy may describe validation conditions, but it does not resolve unequal pilot outcomes in this application context.
  4. The model needs context-specific subgroup testing and monitoring. ✓
    Contrasting observations show that intended-use evidence is incomplete and require subgroup evaluation, contextual monitoring, and documented corrective decisions.
The trap
Separate what validation demonstrated from what deployment evidence still needs to establish.

How to remember it

The model card supports limited validation claims; unequal pilot outcomes require contextual subgroup testing and continuing monitoring.

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 — free

More 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.