AI-102 exam dumps

AI-102 practice question 440 of 493

Designing and Implementing a Microsoft Azure AI Solution. Professional level, Microsoft. Free question with the correct answer and a full explanation.

AI-102 Question 440

Select 3

You are a data scientist tasked with deploying an Azure OpenAI model to handle customer support inquiries for your organization. The model must be capable of understanding natural language and generating human-like responses. Additionally, sensitive customer data must remain protected during the interaction, and the deployment must comply with your company's security policies. Which steps should you take to correctly select and deploy the Azure OpenAI model while ensuring security and compliance?

  1. A

    Choose the GPT-4 model from Azure OpenAI and deploy it in a virtual network for secure access.

  2. B

    Enable Azure Private Link to ensure communication with the model endpoint remains within your private network.

  3. C

    Deploy the Azure OpenAI model in a public endpoint to allow open access and simplify management.

  4. D

    Use role-based access control (RBAC) to restrict access to the Azure OpenAI deployment.

  5. E

    Store sensitive data in plain text in API requests to simplify interactions with the model.

Show answer and explanation

Correct answers: A, B, D

Explanation

To securely deploy an Azure OpenAI model for customer support inquiries, you need to select an appropriate model like GPT-4 for natural language tasks, ensure secure access by deploying it in a virtual network and enabling Azure Private Link, and enforce access control using RBAC. Avoid public endpoints for sensitive use cases and ensure sensitive data is protected in transit and storage to comply with security policies.

  • A. Correct.

    This is correct because GPT-4 is well-suited for natural language understanding and generation tasks, and deploying it in a virtual network ensures secure access.

  • B. Correct.

    This is correct because enabling Azure Private Link ensures that communication with the model endpoint remains private and secure, which is critical for sensitive data handling.

  • C. Incorrect.

    This is incorrect because deploying the model on a public endpoint exposes it to potential security risks and does not align with the requirement to comply with security policies.

  • D. Correct.

    This is correct because using role-based access control (RBAC) ensures that only authorized users and applications can access the model, enhancing security.

  • E. Incorrect.

    This is incorrect because storing sensitive data in plain text violates security best practices and could lead to data breaches.

Timed practice exam

Take a AI-102 practice test under exam conditions

40 questions in 120 minutes, drawn from this bank, with a score report and a per-question review when you finish.

Start timed exam