AIP-C01 Practice Test 4 — 25 Questions

Question 1

A healthcare company requires that all generative AI outputs be filtered for PII (Personally Identifiable Information) before being sent to the end user. Which Amazon Bedrock feature effectively manages this compliance requirement?

Choose an answer to begin — your selection is scored in the full session.

Accepted Answer

Amazon Bedrock Guardrails. Guardrails for Amazon Bedrock allows developers to configure PII filtering, profanity filters, and denial topics to ensure compliance.

Answer

AWS CloudTrail logs

Answer

AWS WAF on the API Gateway

Answer

IAM Policy evaluation for Bedrock

Question 2

A developer is creating a test suite for an LLM application. Which THREE of the following metrics are commonly used to evaluate RAG system performance?

Accepted Answer

Answer Relevance. Faithfulness, answer relevance, and context precision are the key components of RAGAS evaluation.

Answer

Model Training Loss

Answer

Inference Latency

Question 3

When using a Knowledge Base, what is the 'Base Prompt' used for?

Accepted Answer

To instruct the model on how to format the answer using retrieved context.. The base prompt serves as the template for the LLM to understand how to use retrieved context to answer the user's question.

Answer

To define the database schema.

Answer

To authenticate the user.

Answer

To store the conversation history.

Question 4

A company is concerned about model provider policy compliance. Where can they find the specific data processing terms for their chosen model?

Accepted Answer

In the AWS Service Terms and individual model provider documentation.. The AWS Service Terms and the Bedrock documentation page specify the data handling and privacy commitments for each model provider.

Answer

In the Amazon Bedrock console 'Model Evaluation' tab.

Answer

By asking the model during an inference session.

Answer

In the Amazon CloudWatch monitoring dashboard.

Question 5

An organization is concerned about 'Model Hallucinations' and needs a way to ground model outputs in trusted company documents. Which service supports this?

Accepted Answer

Amazon Bedrock Knowledge Bases. Amazon Bedrock Knowledge Bases implements Retrieval Augmented Generation (RAG) by connecting foundation models to trusted data sources stored in Amazon S3 or other supported repositories. It chunks, embeds, and indexes documents into a vector store, then retrieves relevant passages at inference time so the model's response is grounded in actual company content rather than parametric memory. This directly reduces hallucinations by supplying factual context in the prompt.

Answer

Amazon SageMaker Feature Store

Answer

AWS Glue Data Catalog

Answer

Amazon DynamoDB

Question 6

To prevent accidental deletion of a Knowledge Base, which feature should be applied?

Accepted Answer

Apply an IAM policy denying DeleteKnowledgeBase. Apply an IAM policy denying DeleteKnowledgeBase. AWS Identity and Access Management (IAM) policies are the primary mechanism for controlling access to AWS actions, including the DeleteKnowledgeBase action for Amazon Bedrock Knowledge Bases. By explicitly denying this action, you prevent any principal (user, role, or service) from deleting the knowledge base, even if they have other permissions that would otherwise allow it. This is a direct, preventive control that aligns with the principle of least privilege.

Answer

Use AWS Backup

Answer

Enable MFA Delete on S3

Answer

Enable S3 Versioning

Question 7

You have a fleet of SageMaker endpoints. You want to implement an automatic cost reduction strategy during off-peak hours for specific endpoints. Which approach is most efficient?

Accepted Answer

Configuring Application Auto Scaling with scheduled actions. Application Auto Scaling for SageMaker endpoints supports scheduled scaling actions that adjust the desired instance count based on a cron or rate expression. This allows you to automatically scale in (reduce instances) during off-peak hours and scale out during peak hours, directly reducing cost without manual intervention. It is the most efficient because it is automated, precise, and integrates natively with SageMaker endpoints.

Answer

Using spot instances for all endpoints

Answer

Deactivating the endpoint manually

Answer

Updating the endpoint configuration

Question 8

When designing an agentic solution, which THREE aspects of security must be considered?

Accepted Answer

IAM least privilege. Option B (IAM least privilege) is correct because an agentic solution typically assumes roles or service accounts to call tools, models, and data sources, so granting only the minimal permissions needed limits blast radius if the agent is manipulated or compromised. Option D (Applying Guardrails for input/output) is correct because guardrails filter and validate prompts, tool invocations, and model responses, mitigating prompt injection, data leakage, and unsafe or off-policy actions. Option E (Encryption of data at rest and in transit) is correct because agent workflows move sensitive context, credentials, and results across networks and stores, so TLS for transit and KMS-backed encryption for storage protect confidentiality and integrity. Option A (Making all API endpoints public) is not appropriate because exposing every endpoint without authentication or authorization dramatically expands the attack surface. Option C (Sharing API keys in the prompt) is not appropriate because embedding secrets in prompts risks leakage through logs, model context, or outputs and violates credential-handling best practice.

Answer

Making all API endpoints public

Answer

Sharing API keys in the prompt

Question 9

A compliance team requires that all prompts containing sensitive data be encrypted. How does Amazon Bedrock handle this during the request?

Accepted Answer

Amazon Bedrock uses TLS 1.2+ for encryption in transit.. Amazon Bedrock encrypts all data in transit using TLS 1.2 or higher, including prompts sent to the API. This is automatic and does not require manual encryption or special encoding by the developer. Therefore, prompts containing sensitive data are encrypted in transit by default.

Answer

Bedrock only supports encrypted prompts if a VPC endpoint is used.

Answer

The prompt must be base64 encoded.

Answer

The developer must manually encrypt the prompt string before calling the API.

Question 10

A firm wants to audit the specific version of a foundation model used for every inference request. How can they achieve this?

Accepted Answer

Enable Bedrock Model Invocation Logging. Bedrock Model Invocation Logging captures each invocation's request and response metadata — including the model ID and, when applicable, the specific model version or inference profile — and delivers it to CloudWatch Logs or S3. This provides the audit trail needed to prove which foundation model version handled each request.

Answer

Query Amazon SageMaker Model Monitor

Answer

Enable CloudTrail logs

Answer

Use AWS Config to track model changes

Question 11

A team is storing sensitive documents for RAG. They need to ensure that the document chunks in the vector database are encrypted at rest. How?

Accepted Answer

Configure the vector database (e.g., OpenSearch) with KMS encryption. Encryption at rest for a vector database is enforced at the storage layer of that database, so configuring the vector store (e.g., Amazon OpenSearch Serverless or Aurora pgvector) with a KMS customer-managed key ensures the indexed embeddings and metadata are encrypted on disk. This is independent of how the source documents are encrypted in S3. KMS CMKs give the team control over key rotation and access policies.

Answer

Use TLS for all connections

Answer

Use Amazon S3-managed encryption

Answer

Encrypt the source S3 bucket only

Question 12

When should a developer choose to fine-tune a model versus using RAG?

Accepted Answer

When you need the model to learn a specific, consistent corporate tone.. RAG is best for grounding in changing data, while fine-tuning is better for changing the model's style, tone, or highly specialized domain language.

Answer

When you have very little data.

Answer

When you need to ensure the model does not hallucinate.

Answer

When the data changes daily.

Question 13

A company is using a third-party model via Bedrock and needs to ensure that the model provider does not use their data to train their base models. What must they do?

Accepted Answer

Understand that AWS does not use customer data to train their foundation models. AWS does not use customer data to train its foundation models. When using third-party models via Amazon Bedrock, the model provider also does not have access to customer data for training their base models. This is a fundamental data privacy commitment from AWS. Therefore, no additional action is required; the correct understanding is that AWS and its model providers do not use customer data for training. Enabling an opt-out setting is not necessary because the default is already no training on customer data.

Answer

Disable all logging

Answer

Enable the 'Opt-out' setting in the Bedrock console

Answer

Encrypt all data with a customer-managed key

Question 14

When using Bedrock Knowledge Bases, which vector database is integrated natively to support managed RAG workflows?

Accepted Answer

Amazon OpenSearch Serverless. Amazon Bedrock Knowledge Bases natively integrates with Amazon OpenSearch Serverless as its managed vector store, allowing RAG workflows to be set up without provisioning or managing vector database infrastructure. OpenSearch Serverless provides the vector search collection type that Bedrock uses to store and retrieve embeddings during retrieval-augmented generation. This native integration means customers can point Bedrock at an OpenSearch Serverless vector index and have chunking, embedding, and retrieval handled automatically.

Answer

Amazon DynamoDB

Answer

Amazon ElastiCache

Answer

Amazon RDS for PostgreSQL

Question 15

You are troubleshooting a SageMaker inference endpoint and need to see the container logs. Which service stores these logs?

Accepted Answer

Amazon CloudWatch Logs. SageMaker inference endpoints run containers on managed EC2 instances, and all stdout/stderr output from those containers is streamed to Amazon CloudWatch Logs under log groups named /aws/sagemaker/Endpoints/<endpoint-name>. This is where you inspect model loading errors, inference failures, and framework-level messages. CloudWatch is the native logging destination for nearly all AWS managed compute services, making it the correct answer here.

Answer

Amazon S3

Answer

Amazon DynamoDB

Answer

AWS CloudTrail

Question 16

Which AWS service is best used to store and version control your prompt templates for Bedrock?

Accepted Answer

Amazon Bedrock Prompt Management. Amazon Bedrock Prompt Management is the correct answer because it is a purpose-built feature that lets you create, store, version, and reuse prompt templates natively within Bedrock. It provides versioning, aliases, and variables so you can iterate on prompts and roll back to previous versions without external tooling. This integrates directly with Bedrock's InvokeModel and agent APIs, making it the most operationally efficient choice for prompt lifecycle management.

Answer

AWS Systems Manager Parameter Store

Answer

AWS CodeCommit

Answer

Amazon S3

Question 17

You are integrating a SageMaker JumpStart model via the InvokeEndpoint API. The model is failing to process images due to a schema mismatch. Which format does the JumpStart image input expect by default for standard vision models?

Accepted Answer

Base64 encoded string in the JSON input field. SageMaker JumpStart vision models (e.g., image classification, object detection) expect the input image to be Base64-encoded and placed in the JSON payload under the appropriate input field (commonly 'image' or 'data'). The InvokeEndpoint API sends JSON, so raw binary cannot be transmitted directly. The model's inference container decodes the Base64 string back to bytes before preprocessing.

Answer

A URL pointing to an S3 object

Answer

Multi-part form data

Answer

Raw binary stream in the request body

Question 18

What is the primary role of an Embedding Model in a RAG pipeline?

Accepted Answer

To convert text into numerical vectors for semantic similarity search.. Embedding models convert text into numerical vectors that represent the semantic meaning of the content for vector search.

Answer

To act as a firewall for the RAG system.

Answer

To compress large files for storage.

Answer

To generate human-readable summaries.

Question 19

Which TWO of the following actions can help mitigate the risk of 'prompt injection' in a Generative AI application?

Accepted Answer

Sanitize user input before passing it to the model. Option D is correct because sanitizing user input before passing it to the model removes or neutralizes embedded instructions, escape characters, and other adversarial payloads that attackers use to hijack the prompt, directly reducing the attack surface for prompt injection. Option E is correct because Amazon Bedrock Guardrails provides configurable input filtering (including denied topics, content filters, and prompt-attack/word filters) that detects and blocks malicious or manipulative prompts before they reach the foundation model, adding a dedicated defense layer against prompt injection. Option A is not correct because disabling the chat interface eliminates the application's functionality rather than mitigating the risk while preserving legitimate use. Option B is not correct because temperature 0.0 only makes model output more deterministic; it does not prevent an injected instruction from being followed. Option C is not correct because using a more powerful foundation model does not inherently provide prompt-injection protection and may even increase susceptibility to sophisticated adversarial prompts.

Answer

Disable the chat interface

Answer

Set the temperature to 0.0

Answer

Use the most powerful foundation model available

Question 20

Which TWO approaches help optimize the costs of using Amazon Bedrock?

Accepted Answer

Purchasing Provisioned Throughput for predictable workloads. Option C is correct because purchasing Provisioned Throughput for predictable workloads lets you reserve model units at a fixed hourly rate, which is more cost-effective than on-demand pricing when usage is steady and high, and it also provides consistent throughput. Option D is correct because selecting a smaller, more efficient model for simple tasks reduces the per-token cost, since Amazon Bedrock charges based on the model used and smaller models are significantly cheaper than larger ones for tasks that don't require advanced capabilities. Option A is incorrect because using the most expensive model for all tasks unnecessarily increases costs when cheaper models could suffice. Option B is incorrect because disabling all logging does not reduce Bedrock inference costs and may violate compliance or operational requirements. Option E is incorrect because running every request as a separate job adds overhead and does not optimize costs compared to batching or using appropriate model selection.

Answer

Using the most expensive model for all tasks

Answer

Disabling all logging

Answer

Running every request as a separate job