Skip to content

Stage 1 · Beginner · F0

Start Here

Your phone is full of models that learned from examples. Here is what they are.

10 lessons · 97 minGentle

Start chapter

About this chapter

You do not need mathematics, code, or any background at all to understand what AI is doing. You need the right ten pictures, in the right order.

This chapter builds them: what a model is and what it is not, why we stopped writing rules and started showing examples, what data really means, how a pattern is found, and why a guess comes with a number attached. Then the piece everything else is made of: a machine that weighs what it hears and answers yes or no. You finish knowing where AI works, where it fails, why it states wrong things so calmly, and exactly where to go next on the map.

What you will be able to do

  1. 1

    A Hundred Models in Your Pocket

    Name the AI already running on your phone, and say in one sentence what all of it has in common.

    10 min
  2. 2

    Rules, or Examples

    Explain the difference between software that follows written rules and software that learns from examples.

    9 min
  3. 3

    Data Is Just Examples

    Say what data means for a model, and explain why who collected it decides what the model learns.

    9 min
  4. 4

    Finding the Pattern

    Describe what a model actually is, and tell finding a pattern apart from memorizing the examples.

    9 min
  5. 5

    A Guess With a Number On It

    Read a model's answer as a set of guesses with shares, and say what a percentage does and does not promise.

    9 min
  6. 6

    Being Wrong, On Purpose

    Describe training as a loop of guess, measure the miss, adjust, and say when to stop.

    11 min
  7. 7

    What the Model Looks At

    Explain what a feature is, and why choosing what to measure can matter more than the model.

    9 min
  8. 8

    The Weighing Machine

    Describe the simplest model as a machine that weighs its inputs, and say which of its parts training changes.

    10 min
  9. 9

    Bigger Data, More Layers

    Explain in plain words why stacking layers and adding data keeps helping, and what that costs.

    9 min
  10. 10

    Where It Helps, Where It Breaks

    Name the two failures you will meet most often, and choose where to go next on the map.

    12 min

Before you start

Nothing to learn first. You can start here.

Keep going

All chapters