Almost everyone arriving at this subject has an image ready: a machine with a face, arriving at some point in the future, doing something to them.
The image is doing work. It puts the subject somewhere else, in a place where nothing is required this week. It also makes the actual systems invisible, because none of them look like that. The thing sorting your inbox has no face. The thing deciding which of six loan applications gets seen first has no face. The thing that put this video rather than that video in front of your child last night has no face, and nobody in the house can name it.
This level replaces the image twice: first with a line you can draw, then with a list. Lesson 0.1 gives you the one question that separates a system whose behaviour a person wrote down from one that inferred it from examples, because an inventory of a category you cannot draw is a list of whatever you already suspected. Lesson 0.2 removes the three pictures that sit on the wrong side of that line: that the system understands what it says, that it looks things up, and that it is a person. Then, for one week, you write down every system that ranked, filtered, scored or predicted something, what it decided, and what it plausibly used to decide. Then you teach the six-item recognition test to one person who has never used a chatbot, and find out which items they get wrong.
Nothing in this level asks you to use a chatbot. What matters is not which page you read first but that the inventory runs alongside the reading from the first week, because that combination is the one with evidence behind it. Explanation on its own has been measured: a short text explaining how a chatbot works and how it fails produced no reduction in students' adoption of wrong recommendations, and increased their rejection of correct ones (Puppart and Aru, 2025). Explanation combined with drills on the learner's own material did change behaviour (Clerc et al., 2026). Exercise 0.A therefore starts on day one, not after the reading.
Level 0 lessons
0 of 6 done
Replace the picture of artificial intelligence as a future robot with a written inventory of the systems that sorted, ranked, scored or predicted something about you this week.
Done against a real organisation you choose, not a case study. The gates below are about what these produce.
01
Exercise 0.A - Seven days of decisions
Keep the log for seven consecutive days. Do not choose a quiet week, and do not restart because a day was unrepresentative.
Record one row for every occasion you can identify where a system ranked, filtered, scored, predicted or matched something: five fields per row. The service, named. What it decided. What information it could plausibly have used. Whether a person reviewed the output, as far as you can tell. And what you would have wanted to ask about it.
At the end of the week, count the rows by category: message and content filtering, recommendation, identity and verification, scoring and eligibility, prediction. Then mark the three rows you would least want removed and the three you understand least.
You are finished when you have at least fifteen rows across five days, a count by category, and the two sets of three.
For a useful record, avoid two common shortcuts. The first is recording only what happened on a screen, which loses the bank's fraud check, the recruitment screen, the call routing, and the camera at the office door. The second is filling it in on the seventh day from memory, which produces a document that agrees with the impression you already had, which is the document you started with.
02
Exercise 0.B - The recognition test, run on someone else
Take the six items from Lesson 0.3: fitness trackers, customer service chatbots, product recommendations, face recognition on security cameras, music playlist recommendations, email spam filtering. Ask one person who has never used a chatbot, for each item, whether artificial intelligence is involved.
Record their six answers before you say anything. Then record which items they got wrong, what reason they gave, and what they said when you told them the answer. Their reasons matter more than their score.
Then do the harder half. Ask them to name one decision made about them in the past month that they did not understand. Write down what they say verbatim.
For a useful record, do not correct them as you go, because you would then record your own answers. Use someone who has not used a chatbot so you can see what the items look like without that frame.
Level 0 gates
0 of 9 ticked
What you should be able to do after this level. Tick each one when it is true. Nobody marks these. Sign in to tick them.
You have seven days of logged decisions, at least fifteen rows, each naming the service, what it decided, what it could plausibly have used, and whether a person reviewed it.
Not ticked
You can state, from the log rather than from memory, the category that produced the most rows and the three rows whose inputs you cannot name.
Not ticked
You have run the six item recognition test on one person who has never used a chatbot, and recorded their answers, their reasons, and the item they were most confident and wrong about.
Not ticked
You can put a named service you actually use on either side of the written-or-learned line, say which part of it is which, and say who could change each part.
Evidence that shows it: Attach the artefact, or attest with a date if it lives somewhere you cannot upload from.
Not ticked
You can take a sentence in which somebody says a chatbot knew, found or looked up something, and restate what the system actually did, naming which of the three pictures the sentence is carrying.
Not ticked
You can say what a recommender demonstrably does to what you are shown, and what the evidence does not establish about what you believe.
Not ticked
You do not use the phrase "the system decided" as an account of a decision, and can name who accepted the output instead.
Not ticked
You have fixed your task, your subject, your standing exposure and your two named people in writing, with the date, and produced the three test statements about your subject.
Not ticked
You have run the ten statement check once and recorded the date and the lines you scored two or below.
Write down the four elements from the grounding page: your real task with its reader and date, the subject you know well, your standing exposure, and your two named people. Then test the subject: write three statements about it, one true, one false, and one that most people outside the subject would believe but that is wrong. If you cannot produce the third, the subject is not specific enough and you should choose another.
Run the ten statement check and record the date. Score against what you could show someone, not against what you intend.
Choose a local, specific subject rather than a general field. "Education" is not a subject you know better than a chatbot. "The fee structure and examination timetable at the school where I teach" is.
