Retraining may reinforce under-investigation because: What risk arises specifically from the label-generation
Historical allocation affects the labels, allowing retraining to reproduce and reinforce under-investigation.
The question
A public-benefit agency retrains a prioritization model using past investigation outcomes. Staff historically investigated fewer cases from one neighborhood, so those outcomes reflect unequal attention rather than a neutral measure of need. The agency proposes checking only aggregate accuracy. What risk arises specifically from the label-generation process?
Preparing for AIGP? Take the free 5-min readiness quiz →
- Retraining may reinforce under-investigation because historical outcomes encode unequal attention. ✓When labels reflect which cases staff previously investigated, the model can learn and repeat that allocation pattern rather than predict underlying need.
- The agency may need additional representative testing before deployment.Contextual testing is appropriate, but this option does not state how historically selective labels can reproduce under-investigation.
- The agency may need to document the historical data source before retraining.Documentation is useful, but it does not explain how selective investigation can become a repeated allocation pattern in the model.
- Aggregate accuracy can conceal different error rates across neighborhoods.This is a valid evaluation limitation, but it does not identify the risk created by training on outcomes shaped by unequal investigation attention.
The trap
Ask how labels were generated; past decisions may encode unequal attention rather than underlying need. How to remember it
Historical allocation affects the labels, allowing retraining to reproduce and reinforce under-investigation.
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
- The team may miss recurring failure patterns because: What mitigation gap remains? →
- No internal owner may act on alerts or coordinate: What risk remains? →
- Retirement may disrupt the scheduling dashboard because: What risk remains? →
- 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.
