Question 1

Generative AI is a type of artificial intelligence that can ____

🟩 Generate text, images, or other data using generative models.
❌ Generate responses on real-time data.
❌ Perform complex calculations and mathematical operations.
❌ Make predictions about future events.

Explanation:

Generative AI creates new content such as text, images, audio, or code. It does not inherently rely on real-time data or perform heavy numerical calculations.

---

Question 2

How can you avoid hallucinations when using an LLM?

🟩 Train the model on a clean, curated dataset.
❌ Train the model to operate with very few constraints.
❌ Train the LLM on a smaller dataset.
❌ Do not add any context to your prompts.

Explanation:

Hallucinations occur when the model creates incorrect information. High-quality training data and good grounding reduce hallucinations.

---

Question 3

Which scenario would most likely benefit from few-shot prompting?

🟩 Getting a model to convert a list of names into a specific JSON format.
❌ Summarizing a document in pirate style.
❌ Asking for a short story.
❌ Asking for the capital of France.

Explanation:

Few-shot prompting is useful when you need the model to learn a pattern or follow a structured output format. JSON formatting fits this perfectly.

---

Question 4

Which is NOT a good practice for constructing prompts?

❌ Vary your prompts.
🟩 Include long and detailed sentences.
❌ Keep it short and simple.
❌ Avoid ambiguity.

Explanation:

Prompts should be clear and concise. Long, overly detailed sentences introduce confusion and worsen output quality.

---

🧾 Summary Table

---