Start Here: How AI Systems Actually Work

Anyone — no technical background needed7 steps · ~43 min

Build an intuition for what's inside a modern AI assistant, one layer at a time, and see your first thing-goes-wrong story.

By the end: You can explain what's inside a modern AI assistant, why it sometimes makes things up, and how a hidden instruction can hijack one.

0/2 scenarios completed

Start the path →

1
🗺️Conversational Assistant
Reference architecture
Meet the simplest assistant: you, a prompt, a model.
2
🗺️RAG Knowledge Assistant
Reference architecture
Give it a library to look things up in.
3
🌀The Refund That Never Existed~8 min
Interactive scenario · beginner
Watch a confident chatbot invent a policy — and who pays.
4
🗺️Tool-Using Agent
Reference architecture
Now let it take actions in the real world.
5
📧The Email That Gave Orders~10 min
Interactive scenario · intermediate
See how a hidden message in a document hijacks an agent.
6
🧪Sandbox Playground
Build a system, watch risks activate
Build your own system in the Sandbox — watch risks light up as you wire it.
7
📖Glossary
Skim the glossary to lock in the vocabulary.

Next path

AI Risk Practitioner Track →

Work through the OWASP-aligned threat landscape with hands-on incidents, mapping each to root causes, controls, and frameworks (OWASP, ATLAS, NIST AI RMF).