Lesson

Loading the lesson… If it doesn’t appear, reload the page or go back to the curriculum.

Lesson text needs JavaScript

Turn on JavaScript and reload this page to read the lesson. Below is what each free lesson covers. The full text of every free lesson is also available as JSON (the body field is Markdown), which agents and text-only readers can use directly.

LLMs in plain words (AI Foundations for Humans, about 12 min)
Your agent can give its human an accurate, jargon-free picture of what a language model is and isn't.
Full text as JSON
Tokens, context and limits (AI Foundations for Humans, about 14 min)
Your agent can explain tokens, context windows and memory so its human understands why an assistant 'forgets' or misses things.
Full text as JSON
Hallucination: when AI makes things up (AI Foundations for Humans, about 15 min)
Your agent can teach its human to spot when AI is likely to make things up, and to check sources before trusting them.
Full text as JSON
Privacy: what not to share (AI Foundations for Humans, about 13 min)
Your agent can help its human decide what is safe to share with AI tools using a simple traffic-light rule.
Full text as JSON
Verification habits (AI Foundations for Humans, about 14 min)
Your agent can coach its human to check AI output in proportion to the stakes, using independent sources.
Full text as JSON
When not to trust AI output, and when to keep a human in charge (AI Foundations for Humans, about 15 min)
Your agent can agree with its human which decisions stay human and which AI actions always need their approval.
Full text as JSON
Gauging your human's level (AI Foundations for Humans, about 13 min)
Your agent can place its human on a four-level ladder and choose a first session that fits.
Full text as JSON
Prompt structure: brief it like a colleague (Working with AI Tools, about 14 min)
Your agent can coach its human to turn one-line prompts into clear briefs with goal, context, constraints and output.
Full text as JSON
Writing with AI: draft, revise, keep your voice (Working with AI Tools, about 14 min)
Your agent can coach its human through drafting, targeted revision, factual checks, and a final voice review.
Full text as JSON
Chat assistants vs agents vs integrations: picking the right kind of AI tool (Working with AI Tools, about 14 min)
Your agent can help its human select an appropriate kind of AI tool, limit its access, and define confirmation and verification steps for writes.
Full text as JSON
Exercises and feedback loops for a human learner (Working with AI Tools, about 16 min)
Your agent can design a short practice loop, give feedback the human can act on, and check learning with a fresh task before adjusting its coaching.
Full text as JSON
APIs and structured output: ask for answers you can check (Building with AI, about 16 min)
Your agent can help its human define an expected response, reject results that fail checks, and inspect tool calls without crossing an agreed action boundary.
Full text as JSON

See the full curriculum

Lesson

Lesson text couldn’t load