AIF-C01 Question Bank · Latest 50 Questions Free Preview (Page 3/5)
Questions 21–30 of the latest 50 AIF-C01 questions (5 pages total), with answers and explanations, no login required. Practice all 435 questions online.
Question 21 · No. 406 · Single choice
A company is using a generative AI (GenAI) system to generate medical summaries from patient records. The system sometimes produces inaccurate information that appears credible. Which strategy will mitigate the issue?
- A. Implement strict input validation and output formatting rules.
- B. Implement Retrieval Augmented Generation (RAG) to ground responses.
- C. Use smaller, more transparent models for medical applications.
- D. Increase computational resources and model size.
Answer: B
Explanation: The issue described is AI 'hallucination' - when GenAI produces inaccurate but seemingly credible information. Retrieval Augmented Generation (RAG) is the most effective strategy to mitigate this because it grounds the AI's responses in verified, retrieved source documents (such as the actual patient records), reducing the likelihood of fabricated information. Option A only addresses input/output formatting, not factual accuracy. Option C is incorrect because smaller models are typically more prone to hallucinations. Option D (increasing model size) does not solve the hallucination problem, as even large models can produce fabricated content.
Question 22 · No. 407 · Multiple choice
A company is building an agentic AI application by using Model Context Protocol (MCP) servers with Amazon Q Developer CLI. The company wants to follow MCP security practices. Which solutions will meet these requirements? (Choose two.)
- A. Keep MCP servers and the Amazon Q Developer CLI updated.
- B. Monitor MCP logs for normal successful activity.
- C. Use plain text files to store sensitive configurations.
- D. Automatically approve all tools that are provided by MCP servers.
- E. Install servers only from trusted sources.
Answer: A, E
Explanation: Option A (Keep MCP servers and the Amazon Q Developer CLI updated) is a correct security practice because updates often contain important security patches that fix vulnerabilities. Option E (Install servers only from trusted sources) is also correct because installing from untrusted sources can introduce malicious code. Option B is incorrect because you should monitor for abnormal or suspicious activity, not normal successful activity. Option C is incorrect because sensitive configurations should never be stored in plain text files - they should be encrypted. Option D is incorrect because automatically approving all tools is a security risk; tools should be reviewed and approved selectively.
Question 23 · No. 408 · Single choice
A company wants to establish a data governance framework for its AI system. The framework must give the company the ability to monitor when data is generated, updated, and scheduled for deletion. Which data governance component meets these requirements?
- A. Data residency
- B. Data lifecycle management
- C. Data observability
- D. Data logging
Answer: B
Explanation: The question asks about a data governance component that provides the ability to monitor when data is generated, updated, and scheduled for deletion. Data lifecycle management specifically governs data throughout its entire lifecycle, including creation (generation), modification (updates), and disposal (deletion scheduling). Data residency refers to where data is stored geographically. Data observability focuses on data quality and pipeline health. Data logging simply records events. Only Data lifecycle management covers all three requirements mentioned: generation, updating, and scheduled deletion.
Question 24 · No. 409 · Single choice
A company is training a foundation model (FM). The company wants the FM training data to be well managed, secure, and compliant throughout the data lifecycle. The data must follow governance practices. Which option is an AWS data governance strategy for managing data across the data lifecycle?
- A. Allow unrestricted access to all data for every employee.
- B. Store data indefinitely without classification or retention policies.
- C. Rely solely on manual reviews for data quality and compliance.
- D. Establish data lifecycle policies, access controls, logging, and monitoring.
Answer: D
Explanation: Option D is the correct choice because it encompasses all the essential elements of a comprehensive AWS data governance strategy across the data lifecycle. Establishing data lifecycle policies ensures data is managed from creation to deletion according to defined rules. Access controls provide security by restricting who can access the data based on permissions. Logging and monitoring enable observability, auditability, and compliance verification. Options A, B, and C all represent poor governance practices: unrestricted access violates security principles, indefinite storage without policies leads to compliance issues and unnecessary costs, and relying solely on manual reviews is error-prone and inefficient compared to automated governance solutions like AWS Lake Formation, AWS IAM, AWS CloudTrail, and Amazon S3 lifecycle policies.
Question 25 · No. 410 · Single choice
A company needs to document an ML model's purpose, owners, and limitations. Which AWS service or feature provides this capability?
- A. Amazon Bedrock
- B. AWS AI Service Cards
- C. AWS Prescriptive Guidance
- D. Amazon SageMaker Model Cards
Answer: D
Explanation: Amazon SageMaker Model Cards is specifically designed to document machine learning model information, including the model's purpose, owners, intended uses, risk ratings, and limitations. It provides a centralized place to track model metadata throughout its lifecycle, which directly matches the company's requirement to document the ML model's purpose, owners, and limitations. Amazon Bedrock is for building generative AI applications, AWS AI Service Cards document AWS's own AI services (not customer models), and AWS Prescriptive Guidance provides general architectural guidance.
Question 26 · No. 411 · Multiple choice
A company needs to evaluate AI-generated summaries against reference text. The company must measure semantic similarity and coverage of key information. Which metrics meet these requirements? (Choose two.)
- A. F1 score
- B. BERTScore
- C. Bilingual Evaluation Understudy (BLEU) score
- D. Accuracy
- E. Recall-Oriented Understudy for Gisting Evaluation (ROUGE)
Answer: B, E
Explanation: BERTScore is chosen because it measures semantic similarity by leveraging contextual embeddings from BERT, which captures the meaning of text rather than just surface-level word matches. This directly addresses the requirement of measuring semantic similarity between AI-generated summaries and reference text. ROUGE is chosen because it focuses on recall-oriented metrics (such as ROUGE-N and ROUGE-L), which measure how much of the key information from the reference text is covered in the generated summary. This directly addresses the requirement of measuring coverage of key information.
Question 27 · No. 412 · Multiple choice
A human resources company is building an AI/ML solution to screen resumes and select relevant applicants for various job roles. Which AI methodologies will help ensure impartial treatment for all demographic groups? (Choose two.)
- A. Modularity
- B. Fairness
- C. Explainability
- D. Secrecy
- E. Underfitting
Answer: B, C
Explanation: To ensure impartial treatment for all demographic groups in an AI/ML resume screening system, we need: B. Fairness - This directly addresses the need for unbiased treatment across different demographic groups. Fairness in AI involves identifying and mitigating biases that could lead to discriminatory outcomes. C. Explainability - This helps stakeholders understand how the AI makes decisions, which is crucial for identifying potential biases and ensuring the system treats all groups impartially. Explainability allows for auditing and transparency in the decision-making process. The other options are not directly related to ensuring impartial treatment: Modularity is about software design, Secrecy would hinder transparency, and Underfitting is a modeling issue not related to fairness.
Question 28 · No. 413 · Single choice
An AI practitioner wants to quickly deploy and fine-tune foundation models (FMs) for generative AI (GenAI) tasks. The AI practitioner needs access to pre-trained models, solution templates, and example notebooks that can accelerate the development process. Which AWS feature or resource meets these requirements?
- A. Amazon SageMaker Studio
- B. Amazon SageMaker JumpStart
- C. Amazon Bedrock model catalog
- D. Amazon SageMaker Canvas
Answer: B
Explanation: Amazon SageMaker JumpStart is the correct answer because it is specifically designed to help AI practitioners quickly deploy and fine-tune foundation models. It provides a centralized hub with pre-trained models, solution templates, and example notebooks, which directly matches the requirements stated in the question. JumpStart accelerates the development process by offering ready-to-use resources that can be easily customized for specific generative AI tasks. Option A (SageMaker Studio) is an IDE for ML development but doesn't focus on pre-trained models and templates. Option C (Bedrock model catalog) primarily offers foundation models but lacks the comprehensive templates and notebooks. Option D (SageMaker Canvas) is a no-code ML tool aimed at business analysts, not practitioners looking for advanced model customization.
Question 29 · No. 414 · Single choice
A company wants to use foundation models (FMs) for multiple use cases, including content summarization, code generation, and customer service AI assistants. Which solution will implement FMs MOST cost-effectively?
- A. Select the smallest model that meets the performance requirements for each specific use case.
- B. Use the largest and most capable model available for all tasks to ensure quality.
- C. Build and train a custom FM for each individual use case.
- D. Use batch processing for all inference requests to reduce the costs and latency for each request.
Answer: A
Explanation: The most cost-effective approach is to select the smallest model that meets the performance requirements for each specific use case. Different use cases have different complexity levels - content summarization, code generation, and customer service AI assistants each have unique requirements. By matching the model size to the specific task requirements, you optimize costs while maintaining acceptable performance. Option B (largest model for all tasks) is the most expensive. Option C (custom FMs for each use case) involves extremely high costs for training and infrastructure. Option D (batch processing) reduces costs but increases latency, which is unacceptable for real-time customer service interactions.
Question 30 · No. 415 · Single choice
A company that runs AI workloads on Amazon Bedrock needs to evaluate and document compliance with regulatory standards. Which AWS service or feature meets these requirements?
- A. AWS Audit Manager
- B. Amazon SageMaker Model Cards
- C. Amazon SageMaker Model Monitor
- D. AWS Artifact
Answer: A
Explanation: AWS Audit Manager is the correct answer because it is specifically designed to help companies evaluate and document compliance with regulatory standards. It automates evidence collection for audits and provides a framework for continuous compliance assessment. While AWS Artifact provides access to compliance reports, it doesn't evaluate compliance. SageMaker Model Cards and Model Monitor are focused on ML model documentation and monitoring, not regulatory compliance evaluation.