AWS AI Practitioner (AIF-C01) · 14% practice allocation

AWS AIF-C01: Guidelines for Responsible AI Practice Questions

5 free sample questions from the Guidelines for Responsible AI domain, with answers and explanations. The full AIF-C01 bank has 60 questions across 5 domains. Questions are scenario-based and written from the official AWS exam guide.

1.

A loan-approval model is found to reject applicants from a particular demographic at a much higher rate despite similar financial profiles. Which responsible AI concern does this represent?

  • A. Token cost
  • B. Model file size
  • C. Latency
  • D. Bias / fairness✓ Correct
Explanation

Systematically different outcomes for a demographic group despite similar profiles is a bias/fairness problem — a core responsible AI concern. Latency (C), token cost (A), and file size (B) are performance/cost attributes, not fairness issues.

2.

Which of the following are recognized DIMENSIONS of responsible AI? (Select 2)

  • A. Maximum GPU count
  • B. Number of S3 buckets
  • C. CDN cache hit ratio
  • D. Transparency and explainability✓ Correct
  • E. Fairness and bias mitigation✓ Correct
Explanation

Fairness/bias mitigation and transparency/explainability are core responsible-AI dimensions (alongside safety, robustness, privacy, and governance). GPU count (A), bucket count (B), and cache hit ratio (C) are infrastructure metrics, not responsible-AI dimensions.

3.

Which document provides standardized information about a model's intended use, performance, limitations, and considerations to promote transparency?

  • A. A model card✓ Correct
  • B. An IAM policy
  • C. A VPC route table
  • D. A billing invoice
Explanation

A model card documents a model's intended use, performance characteristics, limitations, and ethical considerations — a transparency artifact (Amazon SageMaker offers Model Cards). IAM policies (B), route tables (C), and invoices (D) serve security, networking, and billing purposes.

4.

Which Amazon SageMaker feature helps DETECT bias in datasets and models and provides explainability for predictions?

  • A. Amazon SageMaker Clarify✓ Correct
  • B. Amazon SageMaker Ground Truth
  • C. Amazon SageMaker Neo
  • D. Amazon SageMaker Feature Store
Explanation

SageMaker Clarify detects bias across the ML lifecycle and explains model predictions (feature importance). Ground Truth (B) labels data, Neo (C) optimizes models for hardware, and Feature Store (D) manages features — none is the bias/explainability tool.

5.

A training dataset for a hiring model contains mostly examples from one gender, causing the model to underperform for others. What is the BEST responsible-AI mitigation?

  • A. Deploy the model as-is to save time
  • B. Address the data imbalance (collect more representative data / rebalance) and measure bias before deployment✓ Correct
  • C. Increase the model's temperature
  • D. Reduce the number of tokens per request
Explanation

Bias often originates in unrepresentative data; the fix is to improve dataset representativeness and measure/mitigate bias (e.g., with SageMaker Clarify) before deployment. Deploying as-is (A) perpetuates harm, and temperature (C) or token count (D) have nothing to do with dataset bias.

Test yourself across every AIF-C01 domain

The free AIF-C01 practice test mixes about 20 questions weighted like the exam. The full 60-question bank adds instant grading, flashcards and a timed mock exam.

Take the free AIF-C01 practice testPractice the full bank

Study this domain

Guidelines for Responsible AI study guide (AIF-C01) →

Other AIF-C01 domains

← All AWS AI Practitioner study resources