AIF-C01 Guidelines for Responsible AI Practice Question
A retail bank trained a loan-approval model using Amazon SageMaker. Before deployment, the compliance team asks the ML engineer to produce a report that shows, for each input feature, how strongly its values influence the model's predictions, so reviewers can confirm the model is not making decisions based on a protected attribute such as postal code. Which SageMaker Clarify capability should the engineer use to generate this feature-attribution report?
⚠ Common exam trap
The trap here is assuming any SageMaker Clarify or monitoring output reveals feature influence, when only SHAP-based feature attribution quantifies how each input affects an individual prediction.
Answer choices
Why each option matters
Answer the question above first, then reveal the full breakdown to understand why each option is right or wrong.
Correct answer & explanation
✓
SHAP (Shapley Additive exPlanations) analysis via SageMaker Clarify
Feature attribution is needed to show which inputs drive predictions. SageMaker Clarify's SHAP analysis produces per-feature contribution values that reviewers can inspect to confirm the loan model is not leaning on a protected or proxy attribute. Drift detection, dataset-level bias metrics, and Trusted Advisor do not attribute model behavior to individual input features.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✓
SHAP (Shapley Additive exPlanations) analysis via SageMaker Clarify
Why this is correct
SageMaker Clarify computes SHAP values that assign each input feature a contribution to a prediction, producing global and local feature-attribution reports. This directly answers the compliance requirement by quantifying how strongly each feature, including postal code, influences outcomes, letting reviewers detect reliance on protected or proxy attributes before the model goes into production.
- ✗
AWS Trusted Advisor security and fault-tolerance checks
Why it's wrong here
Trusted Advisor inspects AWS account resources for cost, security, fault tolerance, performance, and service quota issues. It has no visibility into a SageMaker model's internal decision logic and cannot produce feature-attribution values for inputs like postal code, so it is entirely unrelated to the fairness-reporting task the bank requires.
- ✗
Amazon SageMaker Model Monitor data drift detection
Why it's wrong here
Model Monitor data drift detection compares live inference input distributions against a baseline to flag statistical shifts over time. It does not attribute a prediction to individual features or reveal which input drives a decision. It cannot tell reviewers whether postal code influences approvals, so it does not satisfy the feature-attribution reporting requirement the compliance team described.
- ✗
Amazon SageMaker Clarify pre-training bias metrics
Why it's wrong here
Pre-training bias metrics such as class imbalance and difference in positive proportions in labels are computed on the dataset and its labels, not on model behavior. They describe label skew across facets but do not quantify how an individual input feature affects model predictions, so they cannot show whether postal code drives approval decisions.
Go deeper
Related to this question
About these practice questions
Courseiva writes every AIF-C01 question from scratch — 862 in total, each with an explanation and a wrong-answer breakdown. None are copied from real exams or dumps. Learn why practice questions differ from exam dumps →
JA
Written and reviewed by Johnson Ajibi, MSc IT Security
Senior Network & Security Engineer · founder of Courseiva
Last reviewed September 2026 · checked against the official Amazon Web Services exam blueprint
This AIF-C01 practice question is part of Courseiva's free Amazon Web Services certification practice question bank. Courseiva provides original exam-style practice questions with explanations, topic-based practice, mock exams, readiness tracking, and study analytics to help learners prepare for the AIF-C01 exam.