A company has developed an ML model to approve or reject loan applications. The model’s decision-making process must be transparent and explainable to comply with regulatory requirements. The company must document the decision-making process for audit purposes.
Which solution will meet these requirements?
A. Amazon Textract
B. Amazon SageMaker Model Card
C. AWS Cloud Formation
D. Amazon Comprehend
Show Answer
Correct Answer: B
Explanation: Amazon SageMaker Model Cards are designed to document model details, intended use, training and evaluation information, explainability, fairness considerations, limitations, and governance metadata. They support regulatory compliance and audit requirements by providing standardized documentation of the model's decision-making context and lifecycle. The other options do not provide ML model governance documentation: Amazon Textract extracts text from documents, AWS CloudFormation provisions infrastructure, and Amazon Comprehend is an NLP service.
Question 205
HOTSPOT
-
Select the correct AI term from the following list for each statement. Each AI term should be selected one time.
Show Answer
Correct Answer: Simulates human problem-solving capabilities → AI
Applies data-driven learning techniques to make predictions → ML
Focuses on processing data through intricate neural networks → Deep learning
Explanation: AI is the broad field of simulating intelligent behavior. Machine learning uses data to learn predictive models. Deep learning is a subset of ML based on multi-layer neural networks.
Question 206
A company deployed a model to production. After 4 months, the model inference quality degraded. The company wants to receive a notification if the model inference quality degrades. The company also wants to ensure that the problem does not happen again.
Which solution will meet these requirements?
A. Retrain the model. Monitor model drift by using Amazon SageMaker Clarify.
B. Retrain the model. Monitor model drift by using Amazon SageMaker Model Monitor.
C. Build a new model. Monitor model drift by using Amazon SageMaker Feature Store.
D. Build a new model. Monitor model drift by using Amazon SageMaker JumpStart.
Show Answer
Correct Answer: B
Explanation: Model inference quality degrading after several months is commonly caused by data or concept drift. Retraining the model restores performance, and Amazon SageMaker Model Monitor is the AWS service designed to detect and monitor data/model quality drift on production endpoints and trigger alerts through CloudWatch/SNS when thresholds are exceeded. SageMaker Clarify focuses on bias and explainability, while Feature Store and JumpStart are not model drift monitoring services.
Question 207
An ecommerce company is deploying a chatbot. The chatbot will give users the ability to ask questions about the company’s products and receive details on users’ orders. The company must implement safeguards for the chatbot to filter harmful content from the input prompts and chatbot responses.
Which AWS feature or resource meets these requirements?
A. Amazon Bedrock Guardrails
B. Amazon Bedrock Agents
C. Amazon Bedrock inference APIs
D. Amazon Bedrock custom models
Show Answer
Correct Answer: A
Explanation: Amazon Bedrock Guardrails are designed to help filter harmful or unsafe content in both user prompts and model responses. They also support content filtering, denied topics, sensitive information protection, and other safety controls. Bedrock Agents orchestrate workflows, inference APIs invoke models, and custom models address model customization rather than safety filtering.
Question 208
A financial company has offices in different countries worldwide. The company requires that all API calls between generative AI applications and foundation models (FM) must not travel across the public internet.
Which AWS service should the company use?
A. AWS PrivateLink
B. Amazon Q
C. Amazon CloudFront
D. AWS CloudTrail
Show Answer
Correct Answer: A
Explanation: AWS PrivateLink provides private connectivity between VPCs and supported AWS services, including services such as Amazon Bedrock via VPC endpoints, ensuring API traffic remains on the AWS network and does not traverse the public internet. Amazon Q is an AI assistant, CloudFront is a CDN, and CloudTrail is for auditing and logging.
Question 209
A design company is using a foundation model (FM) on Amazon Bedrock to generate images for various projects. The company wants to have control over how detailed or abstract each generated image appears
Which model parameter should the company modify?
A. Model checkpoint
B. Batch size
C. Generation step
D. Token length
Show Answer
Correct Answer: C
Explanation: Generation steps (inference steps) control how many refinement iterations the image generation model performs. Increasing the number of steps generally produces more detailed, refined images, while fewer steps tend to yield rougher or more abstract outputs. Model checkpoints are saved model states, batch size affects processing/training efficiency, and token length is primarily relevant to text generation.
Question 210
A healthcare company is building an AI solution to predict patient readmission within 30 days of patient discharge. The company has trained a model on historical patient data including medical history, demographics, and treatment specifications, to provide readmission predictions in real time.
Which task describes AI model inference in this scenario?
A. Gather historical patient readmission data.
B. Use appropriate metrics and assess model performance.
C. Use data to identify patient patterns and correlations.
D. Use a trained model to predict patient readmission.
Show Answer
Correct Answer: D
Explanation: AI model inference is the process of using an already trained model to make predictions on new, unseen data. In this scenario, the model has been trained on historical patient data and is being used to generate real-time predictions of 30-day patient readmission. The other options describe data collection (A), model evaluation (B), and model training/data analysis (C).
Question 211
A company needs to apply numerical transformations to a set of images to transpose and rotate the images.
Which solution will meet these requirements in the MOST operationally efficient way?
A. Create a deep neural network by using the images as input.
B. Create an AWS Lambda function to perform the transformations.
C. Use an Amazon Bedrock large language model (LLM) with a high temperature.
D. Use AWS Glue Data Quality to make corrections to each image.
Show Answer
Correct Answer: B
Explanation: The requirement is to perform straightforward numerical image transformations (such as transpose and rotation), which are deterministic image-processing operations rather than AI/ML tasks. An AWS Lambda function is a serverless, operationally efficient way to execute these transformations, scales automatically, and can use standard image-processing libraries. A deep neural network and an Amazon Bedrock LLM are unnecessary for this use case, and AWS Glue Data Quality is intended for data quality validation, not image manipulation.
Question 212
A company is working on a large language model (LLM) and noticed that the LLM’s outputs are not as diverse as expected.
Which parameter should the company adjust?
A. Temperature
B. Batch size
C. Learning rate
D. Optimizer type
Show Answer
Correct Answer: A
Explanation: Temperature controls the randomness of token sampling during text generation. Increasing the temperature generally produces more varied and diverse outputs, while lowering it makes outputs more deterministic. Batch size, learning rate, and optimizer type are training-related parameters and do not directly control inference-time output diversity.
Question 213
A company wants to identify groups for its customers based on the customers’ demographics and buying patterns.
Which algorithm should the company use to meet this requirement?
A. K-nearest neighbors (k-NN)
B. K-means
C. Decision tree
D. Support vector machine
Show Answer
Correct Answer: B
Explanation: K-means is an unsupervised clustering algorithm used to group similar customers based on features such as demographics and purchasing behavior when no labels are provided. The other options (k-NN, decision tree, and SVM) are supervised learning algorithms primarily used for classification or regression.
$19
Get all 443 questions with detailed answers and explanations
Instant download HTML + PDF delivered the moment payment clears.
Secure Stripe checkout we never see or store your card details.
7-day refund if files are defective see our refund policy.
