QuiztudyPREMIUM
🌍 All Study Guides📊 Dashboard📰 Blog💡 About
AWS Certified AI Practitioner • STUDY MODE

Practice Quiz

QUESTION 1 OF 14

A security administrator needs to configure Amazon Bedrock Guardrails to evaluate only a specific section of an input prompt using tags. Where can this specific capability be configured and executed?

A
Exclusively through the AWS SDKCorrect Answer
B
Within the Amazon Bedrock Guardrails management console
C
Directly inside the Amazon Bedrock Playground interface
D
Under the general AWS Management Console settings
Explanation:

Evaluating only a specific section of an input prompt using tags is exclusively supported through the AWS SDK, and is unavailable in the management console, Bedrock Playground, or Bedrock Guardrails console.

QUESTION 2 OF 14

An organization is building a pre-processing pipeline and wants to evaluate user inputs against security guidelines using Amazon Bedrock Guardrails, but wants to perform this check independently without invoking or querying any foundation models. How can they achieve this?

A
By invoking the ApplyGuardrail API directly without querying a foundation model.Correct Answer
B
By specifying the Guardrail ID and version during a standard foundation model inference API call.
C
By deploying a custom word filter inside the Bedrock Playground's offline simulator.
D
By using probabilistic entity detection within the AWS SDK's model-free configuration.
Explanation:

Amazon Bedrock Guardrails can be invoked independently via the ApplyGuardrail API without querying a foundation model. Specifying the Guardrail ID and version during standard inference would invoke the foundation model, which violates the requirement.

QUESTION 3 OF 14

A financial organization is building an application using Amazon Bedrock. They need to protect their application from harmful content or denied topics that might be introduced within code elements, specifically comments, variable and function names, and string literals. Which tier of Amazon Bedrock Guardrails should they use to extend content filters and denied topics protection to these specific code elements?

A
Free tier
B
Standard tierCorrect Answer
C
Basic tier
D
Enterprise tier
Explanation:

According to the official AWS documentation, Amazon Bedrock Guardrails extends detection of undesirable content and denied topics protection to code elements—including comments, variable and function names, and string literals—specifically with the Standard tier.

QUESTION 4 OF 14

A software developer wants to selectively evaluate only a specific section of an input prompt using tags with Amazon Bedrock Guardrails, rather than evaluating the entire prompt. Which interface or tool must the developer use to access this capability?

A
The AWS Command Line Interface (CLI) only
B
The Amazon Bedrock Playground console
C
The AWS SDKCorrect Answer
D
The Amazon Bedrock Guardrails management console
Explanation:

The ability to selectively evaluate only a section of an input prompt using tags is available exclusively through the AWS SDK, and it is not available on the management console, including the Bedrock Playground and the Bedrock Guardrails management console.

QUESTION 5 OF 14

Which of the following correctly describes the capability and coverage of Amazon Bedrock Guardrails when filtering undesirable content and protecting sensitive information?

A
It provides controls across foundation models for user inputs or model responses, but excludes reasoning content blocks.Correct Answer
B
It provides controls exclusively for reasoning content blocks, while ignoring standard user inputs.
C
It provides controls for user inputs and model responses, including reasoning content blocks.
D
It provides controls that filter only user inputs, leaving all model responses completely unmonitored.
Explanation:

Amazon Bedrock Guardrails offers safety and privacy controls to filter undesirable content and protect sensitive information in user inputs or model responses, but it explicitly excludes reasoning content blocks.

QUESTION 6 OF 14

An organization wants to build and scale generative AI applications. Which fully managed AWS service provides secure, enterprise-grade access to high-performing foundation models from leading AI companies?

A
Amazon BedrockCorrect Answer
B
Amazon SageMaker
C
Amazon Rekognition
D
AWS Lambda
Explanation:

According to the AWS documentation, Amazon Bedrock is a fully managed service that provides secure, enterprise-grade access to high-performing foundation models from leading AI companies, enabling you to build and scale generative AI applications.

QUESTION 7 OF 14

A software developer wants to configure Amazon Bedrock Guardrails to selectively evaluate only a specific section of their input prompt by applying tags, rather than evaluating the entire prompt. Which of the following describes the supported interface for implementing this selective evaluation?

A
It is available exclusively through the AWS SDK.Correct Answer
B
It is available only through the Bedrock Playground.
C
It is available only through the AWS Management Console.
D
It is available only through the Bedrock Guardrails management console.
Explanation:

The ability to evaluate only a specific tagged section of the input prompt is a feature of Amazon Bedrock Guardrails that is available exclusively through the AWS SDK. It cannot be configured or used via the AWS Management Console, the Bedrock Playground, or the Bedrock Guardrails management console.

QUESTION 8 OF 14

A quality assurance engineer needs to validate user inputs against predefined safety guidelines using Amazon Bedrock Guardrails. To save on costs and separate testing concerns, they must perform this validation independently without invoking any foundation models. How can this request be met?

A
By calling the Guardrail directly using the ApplyGuardrail APICorrect Answer
B
By configuring a model inference call with a mock model parameter
C
By utilizing the Bedrock Guardrails management console test utility
D
By passing a blank model identifier during standard model inference
Explanation:

Amazon Bedrock Guardrails can be invoked directly during foundation model inference or used independently without calling foundation models by utilizing the ApplyGuardrail API.

QUESTION 9 OF 14

A software development company is building an AI assistant that suggests code. They need to configure Amazon Bedrock Guardrails to extend content filtering and denied topics detection to harmful content introduced within code elements, such as comments, variable names, function names, and string literals. Which tier of Bedrock Guardrails must they choose?

A
Standard tierCorrect Answer
B
Basic tier
C
Developer tier
D
Enterprise tier
Explanation:

The Standard tier of Amazon Bedrock Guardrails extends detection of undesirable content and protection against harmful content to code elements, such as comments, variable and function names, and string literals.

QUESTION 10 OF 14

Amazon Bedrock Guardrails provides safety and privacy controls across various foundation models to detect and filter undesirable content. Which of the following content types is excluded from detection and filtering by Amazon Bedrock Guardrails?

A
User inputs containing prompt attacks
B
Reasoning content blocks in model responsesCorrect Answer
C
Model responses containing hate speech
D
User inputs containing insults
Explanation:

Amazon Bedrock Guardrails helps protect sensitive information and detect/filter undesirable content in user inputs and model responses, but it explicitly excludes reasoning content blocks from this detection and filtering.

QUESTION 11 OF 14

A developer is configuring Amazon Bedrock Guardrails to protect a generative AI application. They need to evaluate only a specific section of the user's input prompt by applying tags. How can this selective evaluation be configured?

A
It can be configured only through the AWS SDK, and is not available on the AWS Management Console.Correct Answer
B
It can be configured only through the AWS Management Console, and is not supported in the AWS SDK.
C
It can be configured through both the AWS Management Console and the AWS SDK.
D
It can only be configured by invoking the ApplyGuardrail API on a fully deployed foundation model.
Explanation:

According to AWS documentation, the ability to selectively evaluate only a section of the input prompt by applying tags is available through the AWS SDK and is not available on the AWS Management Console.

QUESTION 12 OF 14

A software company is building a custom text processing pipeline. They want to use the content filtering and sensitive information masking features of Amazon Bedrock Guardrails to inspect and sanitize user comments before they are stored in a database. However, to minimize latency and control costs, the company wants to run these checks without invoking or querying any Amazon Bedrock foundation models. How can the company meet these requirements?

A
Use the ApplyGuardrail API to invoke Amazon Bedrock Guardrails directly without invoking any foundation models.Correct Answer
B
Configure Bedrock Guardrails to selectively evaluate sections of the prompt using the AWS Management Console.
C
Create a custom model on Amazon Bedrock that acts as a proxy for the ApplyGuardrail API.
D
Apply Bedrock Guardrails, which requires routing all traffic through a foundation model as reasoning content blocks.
Explanation:

Amazon Bedrock Guardrails can be invoked directly through the ApplyGuardrail API without having to invoke any foundation models, allowing users to run guardrail checks independently.

QUESTION 13 OF 14

A developer is implementing safeguards for an AI assistant. Which combination of Amazon Bedrock Guardrails features correctly matches their primary mechanism or purpose as described in the service capabilities?

A
Sensitive information filters use probabilistic detection or custom regex to block/mask data, while contextual grounding checks detect hallucinations by verifying if responses are grounded in the source.Correct Answer
B
Sensitive information filters use content categories like Misconduct, while contextual grounding checks can only be configured through custom regex.
C
Content filters block sensitive information like SSN and Date of Birth, while contextual grounding checks detect prompt attacks.
D
Sensitive information filters protect reasoning content blocks, while contextual grounding checks prevent the invocation of foundation models.
Explanation:

Sensitive information filters block or mask sensitive information (like PII/SSN/DOB) using probabilistic detection or custom regex. Contextual grounding checks detect hallucinations in model responses by identifying if they are factually inaccurate, add new information not grounded in the source, or are irrelevant to the user query.

QUESTION 14 OF 14

A business is looking for an AWS service that can help them easily integrate high-performing foundation models from industry-leading AI companies into their applications. They require a fully managed service that ensures secure, enterprise-grade access to these models. Which AWS service should the business select?

A
Amazon BedrockCorrect Answer
B
Amazon SageMaker
C
Amazon Rekognition
D
Amazon Comprehend
Explanation:

Amazon Bedrock is a fully managed service that delivers secure, enterprise-grade access to high-performing foundation models from industry-leading AI companies, making it the correct choice for this business requirement.

Ready to test your recall?

A security administrator needs to configure Amazon Bedrock Guardrails to evaluate only a specific section of an input prompt using tags. Where can this specific capability be configured and executed?

A
Exclusively through the AWS SDK
B
Within the Amazon Bedrock Guardrails management console
C
Directly inside the Amazon Bedrock Playground interface
D
Under the general AWS Management Console settings

How confident are you in this answer?