PART III — Becoming an Effective AI User
Chapter 10
Claude: Building Something New
From Asking Questions to Making Things
After using ChatGPT and Gemini, my curiosity shifted from asking questions to creating things.
I wanted to see whether AI could help me turn an idea into something I could actually use. In this case, the idea was simple educational games for my granddaughter that could run in a web browser.
I had programmed professionally years ago, so I understood some of what was happening behind the scenes. But I was interested in something much bigger than writing code:
Could AI dramatically shorten the distance between having an idea and having something that works?
AI Can Help Turn an Idea Into a Starting Point
One of the surprises of working with AI is that you don't necessarily have to begin with the technical skill required to build the final product.
You can begin with the idea.
AI can help translate a description into something concrete, then give you something you can examine, test, and improve.
That changes the starting point for a project.
Instead of first asking, "Do I know how to build this?" you can begin asking:
"What would it take to make this idea work?"
The First Version Isn't the Finished Product
This is where the conversation becomes important.
The first result may contain errors. It may misunderstand part of the idea. Something may work differently than expected.
That isn't necessarily a failure.
A useful AI conversation can become a cycle:
describe → create → test → explain what happened → revise
The human remains involved throughout the process.
The AI can produce a starting point. You decide whether that starting point is useful, what needs to change, and when the result is good enough.
One Project Can Lead to Another
My experiment with creating games taught me something that applies far beyond programming.
The same pattern can work with writing, planning, organizing ideas, creating educational material, or building something new.
You don't have to know exactly how the finished project will look before you begin.
You can start with an idea and use the conversation to discover what the idea needs.
What You'll Discover
The book goes further into:
how my experiment with Claude developed;
what happened when the first version wasn't right;
why testing and revision matter;
how AI can help people build things they might otherwise have postponed;
and why the human remains responsible for deciding what the finished result should be.
The larger lesson is about creation, not coding.
Try It Yourself
Think of a small project you've postponed because you weren't sure how to begin.
Don't worry about having the perfect plan.
Describe what you would like to create to an AI assistant and see what it suggests as a starting point.
Then examine the result.
The interesting part may be what you ask next.
Why Read the Book?
Chapter 10 is where the book moves from using AI to get answers toward using AI to make things.
My experience with Claude showed me that the most interesting question wasn't whether AI could generate code.
It was whether AI could help turn an idea into something real through conversation, testing, and revision.
AI can help you start. What you do with the starting point is still up to you.