Confirm the lawful rights to reuse the data for this new: Which action best satisfies data-governance

Data governance requires confirming lawful rights to reuse the data for the intended purpose and fixing quality gaps before any training.

The question

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

1. Proceed with training now and log the consent uncertainty as a residual risk to be revisited after the model has shipped.

   Plausible as risk logging, but shipping on unresolved lawful-basis and quality issues violates the governance gate.
2. Rely on the sheer volume of the records to offset the demographic gaps, since a larger dataset generally improves model accuracy.

   Plausible because size often helps, but volume does not cure representational gaps or the missing lawful basis.
3. Anonymize the records so that the lawful-basis and fit-for-purpose questions supposedly no longer need to be assessed before training begins.

   Plausible since anonymization reduces some risk, but it does not resolve fitness gaps or, if imperfect, the lawful-basis question.
4. Confirm the lawful rights to reuse the data for this new purpose and remediate the demographic gaps before the data is used to train. ✓

   Correct: data governance requires establishing lawful rights for the intended purpose and assessing quality and fit before training proceeds.

How to remember it

Data governance requires confirming lawful rights to reuse the data for the intended purpose and fixing quality gaps before any training.

More Understanding How to Govern AI Development questions

• It evidences compliance and supports risk management: Which rationale most accurately captures the governance →
• Recording data lineage and provenance: Which practice would have prevented this gap? →
• Bias testing across protected groups together: Beyond validation of predictive performance, which combination →
• All 426 Understanding How to Govern AI Development questions →