AIF-C01 practice questions 71–80

1. Question 71 · 417 in bank

   A company has developed an ML model for image classification. The company wants to deploy the model to production so that a web application can use the model. The company needs to implement a solution to host the model and serve predictions without managing any of the underlying infrastructure. Which solution will meet these requirements?

   • AUse Amazon SageMaker Serverless Inference to deploy the model.
   • BUse Amazon CloudFront to deploy the model.
   • CUse Amazon API Gateway to host the model and serve predictions.
   • DUse AWS Batch to host the model and serve predictions.

   Show answer

   Answer: A

   Explanation

   Use Amazon SageMaker Serverless Inference to deploy the model: Amazon SageMaker Serverless Inference allows you to deploy machine learning models in a fully managed, serverless environment. You don't need to manage the underlying infrastructure (such as EC2 instances) to handle predictions. This is ideal for scenarios like yours, where the model needs to be deployed and used by a web application, and scalability and infrastructure management should be abstracted away.

   Source: Adapted from community discussion — Jessiii

   Next question

2. Question 72 · 417 in bank

   An AI company periodically evaluates its systems and processes with the help of independent software vendors (ISVs). The company needs to receive email message notifications when an ISV's compliance reports become available. Which AWS service can the company use to meet this requirement?

   • AAWS Audit Manager
   • BAWS Artifact
   • CAWS Trusted Advisor
   • DAWS Data Exchange

   Show answer

   Answer: B

   Explanation

   B: AWS Artifact Explanation: AWS Artifact is a service that provides on-demand access to AWS compliance reports, including those from independent software vendors (ISVs). AWS Artifact can notify users when new compliance reports are available, ensuring that the company stays updated and can evaluate its systems and processes accordingly. D: AWS Data Exchange: AWS Data Exchange is used for subscribing to and managing third-party data sets. It is not intended for compliance reports or notifications about them. Conclusion: AWS Artifact is the best choice for accessing and receiving notifications about compliance reports from independent software vendors (ISVs).

   Source: Adapted from community discussion — Moon

   References

   • AWS documentation
   • AWS documentation
   • AWS documentation
   • AWS documentation

   Next question

3. Question 73 · 417 in bank

   A company wants to use a large language model (LLM) to develop a conversational agent. The company needs to prevent the LLM from being manipulated with common prompt engineering techniques to perform undesirable actions or expose sensitive information. Which action will reduce these risks?

   • ACreate a prompt template that teaches the LLM to detect attack patterns.
   • BIncrease the temperature parameter on invocation requests to the LLM.
   • CAvoid using LLMs that are not listed in Amazon SageMaker.
   • DDecrease the number of input tokens on invocations of the LLM.

   Show answer

   Answer: A

   Explanation

   Create a prompt template that teaches the LLM to detect attack patterns: By creating a prompt template that is specifically designed to identify and mitigate common attack patterns (e.g., prompt injections or malicious requests), you can make the LLM more robust to manipulation. This can help the LLM recognize when it's being manipulated into performing undesirable actions or exposing sensitive data, and take preventive measures accordingly.

   Source: Adapted from community discussion — Jessiii

   Next question

4. Question 74 · 417 in bank

   A company wants to classify human genes into 20 categories based on gene characteristics. The company needs an ML algorithm to document how the inner mechanism of the model affects the output. Which ML algorithm meets these requirements?

   • ADecision trees
   • BLinear regression
   • CLogistic regression
   • DNeural networks

   Show answer

   Answer: A

   Explanation

   Decision trees provide a transparent and interpretable model. They allow you to understand how decisions are made by following the path of splits at each node, which directly shows the influence of input features (gene characteristics) on the final output (gene category). This helps document and explain how the model arrives at its classifications, satisfying the requirement for transparency regarding the inner mechanism of the model.

   Source: Adapted from community discussion — Jessiii

   Next question

5. Question 75 · 417 in bank

   A company is using the Generative AI Security Scoping Matrix to assess security responsibilities for its solutions. The company has identified four different solution scopes based on the matrix. Which solution scope gives the company the MOST ownership of security responsibilities?

   • AUsing a third-party enterprise application that has embedded generative AI features.
   • BBuilding an application by using an existing third-party generative AI foundation model (FM).
   • CRefining an existing third-party generative AI foundation model (FM) by fine-tuning the model by using data specific to the business.
   • DBuilding and training a generative AI model from scratch by using specific data that a customer owns.

   Show answer

   Answer: D

   Explanation

   D. Building and training a generative AI model from scratch by using specific data that a customer owns gives the company the most ownership of security responsibilities, as they are responsible for all aspects of the model's development and deployment.

   Source: Adapted from community discussion — jove

   References

   • AWS documentation

   Next question

6. Question 76 · 417 in bank

   An AI practitioner has a database of animal photos. The AI practitioner wants to automatically identify and categorize the animals in the photos without manual human effort. Which strategy meets these requirements?

   • AObject detection
   • BAnomaly detection
   • CNamed entity recognition
   • DInpainting

   Show answer

   Answer: A

   Explanation

   A: Object detection Explanation: Object detection is a computer vision technique that identifies and categorizes objects within an image. In this scenario, it can automatically detect animals in the photos and assign them to categories (e.g., "dog," "cat," "bird"). This approach aligns perfectly with the requirement to identify and categorize animals without manual intervention.

   Source: Adapted from community discussion — Moon

   Next question

7. Question 77 · 417 in bank

   A company wants to create an application by using Amazon Bedrock. The company has a limited budget and prefers flexibility without long-term commitment. Which Amazon Bedrock pricing model meets these requirements?

   • AOn-Demand
   • BModel customization
   • CProvisioned Throughput
   • DSpot Instance

   Show answer

   Answer: A

   Explanation

   On-Demand: The On-Demand pricing model allows the company to pay for the compute resources or services they use without requiring a long-term commitment. This model is ideal for companies that need flexibility and have variable usage patterns, as they can scale up or down based on their needs while only paying for what they use. This model provides the most flexibility and is cost-effective for companies with limited budgets.

   Source: Adapted from community discussion — Jessiii

   Next question

8. Question 78 · 417 in bank

   Which AWS service or feature can help an AI development team quickly deploy and consume a foundation model (FM) within the team's VPC?

   • AAmazon Personalize
   • BAmazon SageMaker JumpStart
   • CPartyRock, an Amazon Bedrock Playground
   • DAmazon SageMaker endpoints

   Show answer

   Answer: B

   Explanation

   B. Amazon SageMaker JumpStart Explanation: • Amazon SageMaker JumpStart enables AI teams to quickly deploy and consume foundation models (FMs) within their own VPC. • It provides pre-trained foundation models from AWS and third-party providers, making it easy to fine-tune and integrate them into applications. • VPC Integration: Ensures that models are deployed securely within the team’s AWS environment.

   Source: Adapted from community discussion — SP888

   Next question

9. Question 79 · 417 in bank

   How can companies use large language models (LLMs) securely on Amazon Bedrock?

   • ADesign clear and specific prompts. Configure AWS Identity and Access Management (IAM) roles and policies by using least privilege access.
   • BEnable AWS Audit Manager for automatic model evaluation jobs.
   • CEnable Amazon Bedrock automatic model evaluation jobs.
   • DUse Amazon CloudWatch Logs to make models explainable and to monitor for bias.

   Show answer

   Answer: A

   Explanation

   A. Design clear and specific prompts. Configure AWS Identity and Access Management (IAM) roles and policies by using least privilege access is the best approach for companies to securely use large language models on Amazon Bedrock, as it emphasizes both prompt clarity and access control.

   Source: Adapted from community discussion — jove

   Next question

10. Question 80 · 417 in bank

    A company has terabytes of data in a database that the company can use for business analysis. The company wants to build an AI-based application that can build a SQL query from input text that employees provide. The employees have minimal experience with technology. Which solution meets these requirements?

    • AGenerative pre-trained transformers (GPT)
    • BResidual neural network
    • CSupport vector machine
    • DWaveNet

    Show answer

    Answer: A

    Explanation

    Generative pre-trained transformers (GPT) are powerful natural language processing models that excel in understanding and generating human-like text. In this scenario, a GPT model can be trained or fine-tuned to take natural language input from employees and convert it into structured SQL queries. This makes it accessible for users who may not have technical expertise, allowing them to retrieve the data they need from the database using simple, conversational prompts.

    Source: Adapted from community discussion — jove

    Next question

Keyboard:J Kmove between questionsA–Dselect an answerRshow or hide the answer← →change page

AIF-C01 practice, question 71. 417 questions available.