Fairness and inclusion practices is the correct choice because it is the focus area that directly targets demographic bias and guides actions to reduce approval gaps across age groups.

Fairness and inclusion practices emphasizes measuring disparate impact and auditing features for correlations with sensitive attributes, and it prescribes mitigation techniques such as reweighting, resampling, or constrained optimization to bring outcomes into alignment across cohorts. In cloud settings teams implement bias detection tools and continuous fairness monitoring so parity is checked and maintained over time.

Explainability and transparency improves understanding of why a model made a decision and it can help investigate bias, but explainability alone does not apply corrective measures to ensure equitable outcomes for different demographic groups.

Reliability and robustness strengthens a model against noise and distribution shift and it improves overall stability, but it does not specifically reduce group level disparities or correct demographic bias.

Privacy and data protection protects sensitive information and helps meet regulatory requirements and ethical data handling, but it does not provide the measurement or mitigation methods needed to close approval gaps between age cohorts.

Full AWS Practitioner Certification Question