Identify Benefits and Capabilities of Generative AI Solutions AB-731 Practice Questions (Page 8)

Questions 36–40

1. 36foundation · easy

   What is the primary purpose of retrieval-augmented generation (RAG) in a generative AI solution?

   • ATo compress the model's parameters to reduce storage requirements
   • BTo encrypt the model's outputs for secure transmission
   • CTo generate responses without any external data, relying solely on the model's training
   • DTo retrieve relevant information from an external knowledge source and use it as context for the model's response
   Reveal Solution Discussion Select an answer first
2. 37foundation · easy

   A team is using a generative AI model to draft customer email responses. They notice that the model sometimes produces replies that are too formal or too casual. Which practice would most directly help the team control the tone of the generated emails?

   • ADeploying the model on a faster GPU
   • BSwitching to a larger model version
   • CRefining the prompt with specific tone instructions
   • DIncreasing the model's training data size
   Reveal Solution Discussion Select an answer first
4. 38foundation · easy

   Why is the representativeness of training data important for an AI solution?

   • AIt increases the model's inference speed
   • BIt guarantees the model will be free from bias
   • CIt ensures the model performs well on the diverse range of inputs it will encounter in real-world use
   • DIt reduces the need for model evaluation after deployment
   Reveal Solution Discussion Select an answer first
5. 39foundation · easy

   Which prompt engineering technique is specifically designed to improve a model's performance on complex reasoning tasks by encouraging it to work through a problem step by step?

   • AFew-shot prompting
   • BPrompt chaining
   • CZero-shot prompting
   • DChain-of-thought prompting
   Reveal Solution Discussion Select an answer first
6. 40expert · hard

   A company is deploying a generative AI system that helps employees draft internal communications. The system has access to confidential company information. The company wants to ensure that the AI does not leak confidential information in its outputs. What is the most effective security measure?

   • AUse a smaller language model to reduce the risk of leakage.
   • BRely on the AI's built-in safety features.
   • CImplement output filtering that scans AI responses for confidential data patterns.
   • DDisable the AI's access to confidential data entirely.
   Reveal Solution Discussion Select an answer first