Identify Generative AI Services and Capabilities in Microsoft Azure AI-900 Practice Questions (Page 2)

Questions 6–10

1. 6expert · hard

   A team is choosing between using a prebuilt model from the Azure AI Foundry model catalog and fine-tuning a model for a specialized task. They have limited labeled data but need high accuracy. What should they consider?

   • AFine-tuning always produces higher accuracy than prebuilt models
   • BPrebuilt models cannot be customized at all
   • CFine-tuning may not improve accuracy with limited data; a prebuilt model with prompt engineering may be more effective
   • DThe model catalog only contains prebuilt models that cannot be fine-tuned
   Reveal Solution Discussion Select an answer first
2. 7foundation · easy

   Which feature of Azure OpenAI Service is specifically designed to reduce harmful or inappropriate content in model outputs?

   • AContent filtering
   • BFine-tuning
   • CToken-based billing
   • DModel versioning
   Reveal Solution Discussion Select an answer first
4. 8application · medium

   A developer is building an application that generates product descriptions. The developer wants to ensure the model's responses are safe and appropriate for a general audience, and also wants to track how many tokens each request consumes. Which Azure OpenAI Service feature should they enable?

   • AContent filtering and usage monitoring
   • BPrivate endpoint and VNet integration
   • CAzure AI Search integration
   • DModel fine-tuning and versioning
   Reveal Solution Discussion Select an answer first
5. 9expert · hard

   A developer is using Azure OpenAI Service to build an application that generates code. The developer notices that the model sometimes returns unsafe code snippets. The developer wants to reduce this risk without significantly increasing latency. Which approach should they take?

   • AFine-tune the model on a dataset of safe code examples
   • BUse Azure OpenAI content filtering and add system-level instructions to the prompt
   • CDisable content filtering to improve performance
   • DRoute all responses through a separate content moderation API
   Reveal Solution Discussion Select an answer first
6. 10application · medium

   A company wants to use Azure OpenAI Service to build a summarization tool. They need to ensure the model can handle long documents and provide consistent output. Which Azure OpenAI Service capability should they use?

   • AFine-tuning with domain-specific data
   • BContent filtering
   • CToken-based context window management
   • DModel versioning
   Reveal Solution Discussion Select an answer first