QUESTION 121
Which prompting attack directly exposes the configured behavior of a large language model (LLM)?

A. Prompted persona switches
B. Exploiting friendliness and trust
C. Ignoring the prompt template
D. Extracting the prompt template

Correct Answer: D

QUESTION 122
A company has built a chatbot that can respond to natural language questions with images. The company wants to ensure that the chatbot does not return inappropriate or unwanted images. Which solution will meet these requirements?

A. Implement moderation APIs.
B. Retrain the model with a general public dataset.
C. Perform model validation.
D. Automate user feedback integration.

Correct Answer: A

QUESTION 123
A company makes forecasts each quarter to decide how to optimize operations to meet expected demand. The company uses ML models to make these forecasts. An AI practitioner is writing a report about the trained ML models to provide transparency and explainability to company stakeholders. What should the AI practitioner include in the report to meet the transparency and explainability requirements?

A. Code for model training
B. Partial dependence plots (PDPs)
C. Sample data for training
D. Model convergence tables

Correct Answer: B

QUESTION 124
A medical company wants to develop an Al application that can access structured patient records, extract relevant information, and generate concise summaries. Which solution will meet these requirements?

A. Use Amazon Comprehend Medical to extract relevant medical entities and relationships. Apply rule-based logic to structure and format summaries.
B. Use Amazon Personalize to analyze patient engagement patterns. Integrate the output with a general purpose text summarization tool.
C. Use Amazon Textract to convert scanned documents into digital text. Design a keyword extraction system to generate summaries.
D. Implement Amazon Kendra to provide a searchable index for medical records. Use a template-based system to format summaries.

Correct Answer: A

QUESTION 125
A company has developed a generative text summarization model by using Amazon Bedrock. The company will use Amazon Bedrock automatic model evaluation capabilities. Which metric should the company use to evaluate the accuracy of the model?

A. Area Under the ROC Curve (AUC) score
B. F1 score
C. BERTScore
D. Real world knowledge (RWK) score

Correct Answer: D

QUESTION 126
A hospital is developing an Al system to assist doctors in diagnosing diseases based on patient records and medical images. To comply with regulations, the sensitive patient data must not leave the country the data is located in.

A. Data residency
B. Data quality
C. Data discoverability
D. Data enrichment

Correct Answer: A

QUESTION 127
An airline company wants to build a conversational AI assistant to answer customer questions about flight schedules, booking, and payments. The company wants to use large language models (LLMs) and a knowledge base to create a text-based chatbot interface. Which solution will meet these requirements with the LEAST development effort?

A. Train models on Amazon SageMaker Autopilot.
B. Develop a Retrieval Augmented Generation (RAG) agent by using Amazon Bedrock.
C. Create a Python application by using Amazon Q Developer.
D. Fine-tune models on Amazon SageMaker Jumpstart.

Correct Answer: B

QUESTION 128
A social media company wants to use a large language model (LLM) to summarize messages. The company has chosen a few LLMs that are available on Amazon SageMaker JumpStart. The company wants to compare the generated output toxicity of these models. Which strategy gives the company the ability to evaluate the LLMs with the LEAST operational overhead?

A. Crowd-sourced evaluation
B. Automatic model evaluation
C. Model evaluation with human workers
D. Reinforcement learning from human feedback (RLHF)

Correct Answer: B

QUESTION 129
A company wants to build a generative AI application by using Amazon Bedrock and needs to choose a foundation model (FM). The company wants to know how much information can fit into one prompt. Which consideration will inform the company’s decision?

A. Temperature
B. Context window
C. Batch size
D. Model size

Correct Answer: B

QUESTION 130
A company is introducing a new feature for its application. The feature will refine the style of output messages. The company will fine-tune a large language model (LLM) on Amazon Bedrock to implement the feature. Which type of data does the company need to meet these requirements?

A. Samples of only input messages
B. Samples of only output messages
C. Samples of pairs of input and output messages
D. Separate samples of input and output messages

Correct Answer: C

QUESTION 131
A company deployed an AI/ML solution to help customer service agents respond to frequently asked questions. The questions can change over time. The company wants to give customer service agents the ability to ask questions and receive automatically generated answers to common customer questions. Which strategy will meet these requirements MOST cost-effectively?

A. Fine-tune the model regularly.
B. Train the model by using context data.
C. Pre-train and benchmark the model by using context data.
D. Use Retrieval Augmented Generation (RAG) with prompt engineering techniques.

Correct Answer: D

QUESTION 132
A company wants to keep its foundation model (FM) relevant by using the most recent data. The company wants to implement a model training strategy that includes regular updates to the FM. Which solution meets these requirements?

A. Batch learning
B. Continuous pre-training
C. Static training
D. Latent training

Correct Answer: B

QUESTION 133
A financial company is developing a generative Al application for loan approval decisions. The company needs the application output to be responsible and fair. Which solution meets these requirements?

A. Review the training data to check for biases. Include data from all demographics in the training data.
B. Use a deep learning model with many hidden layers.
C. Keep the model’s decision-making process a secret to protect proprietary algorithms.
D. Continuously monitor the model’s performance on a static test dataset.

Correct Answer: A

QUESTION 134
Which option is a benefit of using Amazon SageMaker Model Cards to document AI models?

A. Providing a visually appealing summary of a model’s capabilities.
B. Standardizing information about a model’s purpose, performance, and limitations.
C. Reducing the overall computational requirements of a model.
D. Physically storing models for archival purposes.

Correct Answer: B

QUESTION 135
A company is testing the security of a foundation model (FM). During testing, the company wants to get around the safety features and make harmful content. Which security technique is this an example of?

A. Fuzzing training data to find vulnerabilities
B. Denial of service (DoS)
C. Penetration testing with authorization
D. Jailbreak

Correct Answer: D

QUESTION 136
A company wants to use Amazon Bedrock. The company needs to review which security aspects the company is responsible for when using Amazon Bedrock. Which security aspect will the company be responsible for?

A. Patching and updating the versions of Amazon Bedrock
B. Protecting the infrastructure that hosts Amazon Bedrock
C. Securing the company’s data in transit and at rest
D. Provisioning Amazon Bedrock within the company network

Correct Answer: C

QUESTION 137
Which prompting technique can protect against prompt injection attacks?

A. Adversarial prompting
B. Zero-shot prompting
C. Least-to-most prompting
D. Chain-of-thought prompting

Correct Answer: A

QUESTION 138
A company is using a large language model (LLM) on Amazon Bedrock to build a chatbot. The chatbot processes customer support requests. To resolve a request, the customer and the chatbot must interact a few times. Which solution gives the LLM the ability to use content from previous customer messages?

A. Turn on model invocation logging to collect messages.
B. Add messages to the model prompt.
C. Use Amazon Personalize to save conversation history.
D. Use Provisioned Throughput for the LLM.

Correct Answer: B

QUESTION 139
An AI practitioner is developing a prompt for an Amazon Titan model. The model is hosted on Amazon Bedrock. The AI practitioner is using the model to solve numerical reasoning challenges. The AI practitioner adds the following phrase to the end of the prompt: “Ask the model to show its work by explaining its reasoning step by step. ” Which prompt engineering technique is the AI practitioner using?

A. Chain-of-thought prompting
B. Prompt injection
C. Few-shot prompting
D. Prompt templating

Correct Answer: A

QUESTION 140
What does an F1 score measure in the context of foundation model (FM) performance?

A. Model precision and recall
B. Model speed in generating responses
C. Financial cost of operating the model
D. Energy efficiency of the model’s computations

Correct Answer: A