How Data Is Stored in an AI Model
A plain-English look at the most misunderstood thing about AI: where its “knowledge” actually lives. Not a folder of files, not a search engine — patterns, not pages. No tech background needed.
Most people picture an AI model as a giant library or a hard drive — a place where everything it ever read is saved, ready to be looked up word for word. That picture is wrong, and the wrong picture quietly leads to real mistakes: over-trusting AI as if it holds the exact source, worrying about privacy in the wrong places, and being baffled when a confident chatbot simply makes something up.
This class clears that up. In short, plain-spoken lessons you'll see what's really inside a model — how words become numbers, how “knowledge” gets baked into millions of tiny dials, and why that design explains both AI's uncanny strengths and its strangest failures. It's the natural next step after Introduction to AI: that class answered what AI is; this one answers how it knows what it knows. A live AI tutor is built right into the page so you can talk or type, in your own language, and test every idea as you go.
What you'll learn
1The Big Misunderstanding
- 1
The Library Picture in Your Head
What “storing data” normally means — and why we assume AI works the same way.
Quiz - 2
A Model Is Not a Folder of Files
Patterns, not pages: the core idea this whole class is built on.
Quiz - 3
Review: The Big Misunderstanding
The one idea to carry forward.
2Turning Words Into Numbers
- 4
Everything Becomes Numbers
Before a model can learn from words, it has to turn them into numbers.
Quiz - 5
Meaning as a Location
How a model places related ideas near each other in “number space.”
Quiz - 6
Review: Turning Words Into Numbers
From letters to locations.
3Where the Knowledge Lives
- 7
Millions of Tiny Dials
Meet the model's parameters — the real home of everything it “knows.”
Quiz - 8
Training Is Tuning the Dials
How reading enormous amounts of text turns into dial settings.
Quiz - 9
The Chef Who Never Reads the Recipe
An everyday picture for how a model holds what it learned.
Quiz - 10
Review: Where the Knowledge Lives
The heart of the class.
4Why This Explains AI's Quirks
- 11
Why AI Makes Things Up
Hallucinations aren't a glitch — they're the flip side of how a model works.
Quiz - 12
When It Quotes Word-for-Word
If models don't store text, how do they sometimes reproduce it exactly?
Quiz - 13
Compression, Not Copying
A model is a lossy summary of patterns — which is its power and its limit.
Quiz - 14
Review: Why This Explains AI's Quirks
One mechanism behind all the surprises.
5What This Means for You
- 15
Where Your Information Actually Goes
Training data versus your chat — two very different things to worry about.
Quiz - 16
The Knowledge Cutoff — and How Tools Fill the Gap
Why a model's knowledge freezes, and how search is bolted back on.
Quiz - 17
Using It Wisely
Turning how a model stores data into everyday good judgment.
Quiz - 18
Review & Next Steps
Where to go from here.
Join the class to read each lesson and take the knowledge checks.