AI Foundations

Generative AI

AI that creates text, images, and code

Some AI sorts things, like a spam filter. Generative AI does something new: it creates. This is the kind of AI you will use to build every app in this course.

What is generative AI?

Generative AI is AI that makes new content when you ask for it. "Generate" just means "create".

Think of the difference between a librarian and a writer. A librarian finds a book that already exists. A writer writes a new page just for you. Generative AI is the writer.

A spam filter answers "spam or not spam?"
Generative AI answers "write me an email", and writes one that never existed before.

What it can create

Generative AI can create many kinds of content. You describe what you want in plain words, and it makes it.

Feature	You ask for	It creates
Text	A birthday message for my sister	A warm, personal message
Images	A logo for a plant shop	A new picture
Code	A to-do list web page	Working HTML and JavaScript
Audio	Read this paragraph aloud	A spoken voice recording

You give a short description. Generative AI creates something new.

In this course, you will mostly use text and code. Here is a code example you can try right now:

Prompt

Create a simple web page with a button.
When I click the button, show a random motivational quote.
Keep it to one HTML file.

Let's break this down:

You described what you want, not how to code it.
The AI writes the code for you.
This is the core idea of building apps without writing code yourself.

What it's good at

Generative AI is a great helper for many everyday tasks:

First drafts of emails, posts, and documents
Explaining hard topics in simple words
Summarizing long text into key points
Brainstorming names, ideas, and plans
Writing code from a plain description

What it's bad at

It also has real weak spots. Knowing them will save you a lot of trouble:

Facts: it can make up facts that sound true. You will learn more about this on the Hallucinations page.
Recent news: it may not know about events after it was trained, unless it can search the web.
Exact math: it can slip on long calculations.
Reading your mind: a vague request gets a vague result.

Treat generative AI like a fast, eager assistant on their first day. The work is often great, but you still check it before you use it.

Common mistakes

Asking too vaguely

Prompt

# Wrong
Write something about my shop.

# Right
Write a 3-sentence welcome message for the home page of my
handmade candle shop. Friendly tone. Mention free shipping over $50.

Trusting every fact

If the AI gives you a statistic, a date, or a quote, check it in a reliable source before you publish it.

Practice

Try these exercises to lock in what you just learned.

Quiz

Not started

Which task is an example of generative AI?

Quiz

Not started

You ask an AI to write a blog post and it includes a surprising statistic. What should you do?

What's next?

The generative AI that writes text and code is powered by something called a large language model. Let's find out what that is.