A reference site, not a course
AI, explained by level.
Nothing to sell you.
Almost everything that ranks for “learn AI” is an affiliate page pointing at a course. This site points at whatever is actually best, says when it's free, and tells you the date it was last checked so you know how much to trust it.
Last verified 11 August 2026 · Next scheduled review: September 2026
Pick where you actually are
The usual mistake is starting in the wrong place. People who just want to use these tools well get pushed into machine-learning theory and quit; people who want to build get handed prompt tips. Three honest tracks:
-
Track 01
Curious
You've maybe typed a question into a chatbot. You don't know what a token is and you shouldn't have to yet. Goal: understand what this thing is, what it's bad at, and why people keep talking about it.
No technical background needed -
Track 02
Practical
You use AI most days and suspect you're using maybe 20% of it. Goal: get genuinely good at the tools in your actual work, and learn enough of the machinery to know why things fail when they fail.
Where most people are, and least served -
Track 03
Builder
You want to put a model inside something you're making. Goal: APIs, tool use, agents, retrieval, and the cost and reliability decisions nobody mentions until your bill arrives.
Some coding assumed
Three rules this site runs on
1. Every page is dated
AI content rots faster than anything else in tech. A guide written eight months ago can be confidently wrong about prices, limits, and which model to use. So every page here carries a last verified stamp, and pages that have gone stale say so rather than quietly misleading you. This is the whole reason the site exists: not to be the biggest, to be the one you can date-check. Every edit and correction is logged, dated, on the changes page, so the promise is checkable rather than just stated. The same method, turned on anything else you read: is what you're reading out of date?
2. Hard numbers live in one place
Context window sizes, prices, and model names change constantly. Rather than bury them in a hundred paragraphs that all rot separately, every number lives on a single Model facts page with its own date. The explainers teach the concept and link to the table for the figure. One page to maintain instead of a hundred to forget.
3. Free first, and no affiliate links
Where a free resource is as good as a paid one, the free one is listed first and the paid one is labelled. There are no affiliate links on this site, which is also why it will never recommend a $600 bootcamp that teaches what a free four-hour course covers better.
What we don't know, we say
Anywhere a figure couldn't be verified against a primary source, you'll see it flagged in orange rather than filled in with something plausible-looking. A confidently wrong number is worse than a visible gap, and “plausible-looking” is exactly what a language model produces when it doesn't know.
The terms everyone drops without explaining
Short, honest explainers for the vocabulary that gets used as though you already know it. Written to actually teach, not to rank.
-
Context windows
Why a model “forgets” mid-conversation, and why bigger isn't automatically better.
-
Tokens
The unit you're actually billed in, and why the ratio everyone quotes is now wrong.
-
Hallucination
Not a bug that will be patched. A property of how these systems work.
-
Training cutoff
Where the model's knowledge stops. Two dates, not one, and it doesn't know its own.
-
Tool use
The model is an author, not an actor. How text becomes actions, and what that risks.
-
Agents
A model in a loop with tools. The rest is marketing, and the loop is the hard part.
-
RAG and retrieval
Answering from your own documents. A search problem wearing an AI costume.
-
Fine-tuning
Changes how a model behaves, not what it knows. The distinction costs people money.
-
Prompt caching
The biggest cost lever most builders have, and the easiest to break by accident.
-
Thinking and reasoning
What "thinking before answering" really is, and why the working shown isn't a confession.
Not sure which track? → Start here and answer three questions. · New to the vocabulary? → Glossary · Want the limits first? → What AI is actually bad at