Generative AI Intermediate Quiz 5

1. What is “model drift” when monitoring a deployed Generative AI system?

2. How are AI-generated images often evaluated for quality in practical settings?

3. In Generative AI, what is a “seed” parameter commonly used to control?

4. What is a potential risk when using Generative AI for sensitive or high-impact topics?

5. What is meant by “post-processing” of Generative AI outputs in a workflow?

6. How can you encourage more creative text from a Generative AI language model?

7. What is the purpose of a “safety filter” in a Generative AI application?

8. Which is a best practice when sharing AI-generated content with other people?

9. In text generation, what does the “top‑p” (nucleus) sampling parameter control?

10. Why do many chat‑style Generative AI systems use a separate “system” instruction?