Lesson 2.1 – Generative AI in Action

Unit 2 · Lesson 2.1 – Generative AI in Action | Knowledge Pillars · AWS AI Explorers

Lesson Objectives

By the end of this lesson you will be able to:

Interact with generative AI models and observe how they respond to different inputs
Compare outputs from the same prompt and identify why results vary
Recognize where AI is helpful and where it produces unreliable or biased content

Try It and Watch

The best way to understand generative AI is to use it. Give it a question, then change your wording and watch the answer change. Small differences in what you ask lead to real differences in what you get back.

💡 Try this: Ask for "ideas for a school project," then ask for "three ideas for a school project that helps new students." Notice how much more useful the second answer is.

Same Prompt, Different Answers

Run the exact same prompt twice and you may get two different answers. That is normal. Generative AI predicts likely responses, and several things shape the result.

🧠

The model

Different AI models answer differently.

📝

Your wording

The exact words and detail you use change the output.

📚

Context

What you include around the request shapes the answer.

🎲

Built-in variety

Models add some variation on purpose, so answers differ.

Helpful or Unreliable?

Knowing where generative AI is strong, and where it is risky, makes you a smarter user and a better builder.

Helpful	Be careful
Brainstorming and drafting	Exact facts and statistics (it can make them up)
Explaining and rephrasing	Anything sensitive or high-stakes
Getting unstuck quickly	Content that may carry bias from its data

🧑‍💻 Activity: Run It Twice

Activity

Compare and judge

Generative AI is not a vending machine that gives one fixed answer. Test that yourself.

Write one prompt. Pick a question you can judge the quality of.
Run it twice. Submit the same prompt two times and read both answers.
Compare. What was the same, and what was different between the two?
Judge it. Which answer was better, and would you trust it without checking? Why?

Lesson Summary

Idea	Why it matters
Try and observe	Different inputs lead to different outputs.
Results vary	The model, your wording, context, and built-in variety all matter.
Know the strengths	Great for brainstorming and drafting.
Know the risks	Weak on exact facts; can be biased or unreliable.
Compare to choose	Comparing outputs helps you pick better prompts and tools.

Vocabulary: tap a card to flip

Tap a card (or press Enter) to reveal its meaning.

Generative AI

AI that creates new text, images, or code

Output

What the AI gives back from your prompt

Vary

To change; the same prompt can produce different answers

Reliable

Trustworthy; AI is more reliable for some tasks than others

Bias

Unfair patterns an AI can reflect from its data

Knowledge Check

1. When you give a generative AI different inputs, it:

2. Why might the SAME prompt give different answers?

3. Generative AI is MOST reliable for:

4. You notice an AI answer seems biased. A good move is:

5. The point of comparing outputs is to:

Lesson Checklist

Used a generative AI tool and observed how it responds to different inputs
Explained why the same prompt can give different answers
Named where AI is helpful and where it is unreliable or biased