AI literacy for IB schools · ages 9–13
We Teach AIbybuilding it.
An eight-session workshop we run in your classroom.
Small teams of students build, train and question their own small AI model — using the same basic ideas as tools such as ChatGPT, scaled down for the classroom — around the unit of inquiry your class is already studying.
8 × 75 min · in your classroom · on-screen and hands-on
Why this, why now
Passengers,
or builders?
The children in your classroom will grow up in an AI-saturated world. The question is whether they enter it as passengers — typing prompts into systems they treat as magic — or as people who understand, at a foundational level, what these machines actually are.
The AI Lab takes the second route. It strips the mystery out of language models by having students build one.
The truth contract
Six promises we make to your students.
It actually runs
Not a simulation: press Train and the model really trains, on real machines. The first strange sentences are your students' own model, talking back.
Every model turns out different
Different choices produce different models. There's no leaderboard and no best build — the variety is exactly what the class compares.
We say what actually happened
Students hear mechanisms, not magic: the algorithm merged the most frequent pair; its settings changed when the prediction missed.
Abilities have causes
A model's behaviour comes from the text it read, how big it is, how long it trained, and how people shaped it afterwards. Students learn to tell those apart.
The lab is the unit
Teams join with a one-time code. No student accounts, no names on inputs — what students write stays on the school's own devices.
Every screen answers "why?"
A child can always ask why a step matters. Plain language first, formal terms behind a "Why?" layer for the curious.
For schools
The practical questions matter too.
The programme brief gives coordinators and school leaders the detail they need to evaluate a pilot, including:
- How does the workshop fit our unit of inquiry and year group?
- What does the school need to provide?
- How are safeguarding, privacy and student access handled?
- What happens before, during and after a pilot?
The workshop · 8 sessions
Eight sessions, built across four acts.
Each session ends by uncovering the problem the next one solves. Students collect text, confront its mess, watch language become numbers, train a real model, teach it to behave — then test it and trace its answers.
The book
Building the Machine
How large language models work — written for a ten-year-old who wanted a real answer.
Building the Machine
"But how does it know all this stuff? How does it understand what I'm asking?"
The human story behind the machine
No one invented the modern language model all at once. Across generations, people solved different problems: how to make a general-purpose computer, gather immense collections of text, represent patterns in language as numbers, build networks that learn, and predict what comes next from context. Building the Machine follows how these separate paths eventually converged in the language models behind tools such as ChatGPT.
The book stands entirely on its own. The AI Lab follows the same scientific sequence and offers some readers a hands-on way to explore it.