AI-102 Question 442
Select 3You are tasked with deploying an Azure OpenAI model for a company's customer service chatbot. The chatbot will use the model to generate natural language responses to user queries. The company requires high scalability, data security for sensitive customer information, and the ability to fine-tune the model. Which steps should you take to meet these requirements?
- A
Deploy the GPT-4 model from Azure OpenAI Studio and integrate it with the chatbot application.
- B
Use Azure OpenAI's fine-tuning capabilities to customize the model with the company's customer service dataset.
- C
Enable private endpoint connectivity to securely access the Azure OpenAI model.
- D
Deploy the model in a region with the lowest latency, regardless of compliance and data residency requirements.
- E
Configure the Azure OpenAI resource to scale automatically based on demand.
Show answer and explanation
Correct answers: A, B, C
Explanation
To meet the company's requirements, you need to deploy and integrate the appropriate Azure OpenAI model, fine-tune it to handle specific customer service scenarios, and ensure secure access using private endpoint connectivity. These steps address the needs for scalability, data security, and customization. Selecting a deployment region should consider compliance and data residency requirements, not just latency, and while auto-scaling is useful, it is not a mandatory step in this scenario.
- A. Correct.
Deploying the GPT-4 model and integrating it with the chatbot application is necessary to provide the required functionality for generating natural language responses.
- B. Correct.
Fine-tuning the model with the company's customer service dataset ensures the chatbot generates contextually relevant and accurate responses.
- C. Correct.
Enabling private endpoint connectivity ensures secure access to the Azure OpenAI model, which is crucial for handling sensitive customer information.
- D. Incorrect.
Ignoring compliance and data residency requirements in favor of latency is not recommended, as it could lead to legal and compliance issues.
- E. Incorrect.
While configuring automatic scaling is beneficial, it is not explicitly required to meet the stated requirements of this scenario.