Your AI path · Review and practice
Your AI path
A map of twenty ARQIV topics, from “What is AI?” to deciding under uncertainty. Read a page, take a three-question check, and the map remembers what you have learned. Five quick questions a day bring back what you are about to forget.
How it works
- Map: topics with what to read first. Green means learned.
- Check: three questions from the page. All three right marks it learned. You can always skip.
- Daily Five: questions you missed come back after 1, 3, 7, 16 and 35 days, a method called spaced repetition.
Hand-written questions from each page. No AI writes or marks anything, and nothing is sent anywhere.
01 · Daily Five
Five questions, then you’re done
Reviews that are due come first, then questions from topics you are ready to start.
Your path
Daily Five and topic map
Answer today’s five, or pick a topic on the map below and check it.
Hand-authored questions, checked against each page.
Class mode: ask the class to vote with hands, then press “Show the answer.” Nothing is saved.
The map
not startedlearninglearnedreview due
Text version. Read the topics in this order; each one builds on the ones before it.
- Map of AI (about 12 minutes)
- How AI works (about 20 minutes)
- Where AI is right now (about 20 minutes)
- Machine learning (about 15 minutes)
- Data and datasets (about 20 minutes)
- Training vs using a model (about 20 minutes)
- Computer vision (about 15 minutes)
- Evaluating AI (about 20 minutes)
- Probability and uncertainty (about 20 minutes)
- Transformers and next-word prediction (about 20 minutes)
- Talking to AI (about 20 minutes)
- Verify It (hallucinations) (about 20 minutes)
- AI that creates (about 20 minutes)
- Embeddings (about 20 minutes)
- Retrieval and RAG (about 20 minutes)
- AI agents (about 25 minutes)
- How agents are built (about 35 minutes)
- Responsible AI (about 25 minutes)
- Work and careers (about 25 minutes)
- Decide and Adapt (about 45 minutes)
02 · For teachers
A five-minute bell-ringer
Use the Daily Five on a projector at the start of class.
How to run it
- Turn on Class mode. It saves nothing on the classroom computer.
- Read each question aloud. Students vote with fingers (1–4) or on mini whiteboards.
- Press Show the answer, read the explanation, and open the page link if the class split.
- Stop after five. The point is a short, regular habit.
What the research says
Quizzing yourself and spacing study out over time were rated the two most useful of ten common study techniques. Documented The mastery check (all three right) follows mastery-learning practice: move on when it is solid, with a way to skip.
No accounts, no class codes, no leaderboards. Each student’s progress stays on their own device.
Honest limits
What this page leaves out
A map and a quiz can get you started, but they do not replace a course.
Keep in mind
Three questions can’t prove you understand a topic; they catch what slipped. The reading-time estimate comes from each page, not from you. Progress lives in this browser, so it does not follow you to another device and disappears if you clear site data. Some pages, such as “Where AI is right now,” change often; their questions will be updated. Every question is hand-written and checked against its page; tell us if one is wrong.
Review status
Written by ARQIV. Initial edition, October 2026. These pages have not been classroom-piloted or independently reviewed yet, and every question and scenario is hand-authored and checked against its sources by ARQIV. Tell us about anything wrong or unclear.
Sources
Where this comes from
Every question links the part of the ARQIV page it comes from.
- John Dunlosky and colleagues, “Improving Students’ Learning With Effective Learning Techniques,” Psychological Science in the Public Interest 14(1), 2013; Association for Psychological Science press release “Which Study Strategies Make the Grade?”.
- ARQIV, “Map of AI” (/ai/map-of-ai/).
- ARQIV, “How AI works” (/ai/how-ai-works/).
- ARQIV, “Where AI is right now” (/ai/where-we-are/).
- ARQIV, “Machine learning” (/ai/machine-learning/).
- ARQIV, “Data and datasets” (/ai/data-and-datasets/).
- ARQIV, “Training vs using a model” (/ai/training-and-inference/).
- ARQIV, “Computer vision” (/ai/computer-vision/).
- ARQIV, “Evaluating AI” (/ai/evaluating-ai/).
- ARQIV, “Probability and uncertainty” (/ai/probability-and-uncertainty/).
- ARQIV, “Transformers and next-word prediction” (/ai/transformers/).
- ARQIV, “Talking to AI” (/ai/talking-to-ai/).
- ARQIV, “Verify It (hallucinations)” (/teachers/verify-lab/).
- ARQIV, “AI that creates” (/ai/creative-ai/).
- ARQIV, “Embeddings” (/ai/embeddings/).
- ARQIV, “Retrieval and RAG” (/ai/retrieval-and-rag/).
- ARQIV, “AI agents” (/ai/agents/).
- ARQIV, “How agents are built” (/ai/how-agents-are-built/).
- ARQIV, “Responsible AI” (/ai/responsible-ai/).
- ARQIV, “Work and careers” (/ai/work-and-careers/).
- ARQIV, “Decide and Adapt” (/future-of-work/decide-and-adapt/).
All links opened and checked October 21, 2026 (accessed October 2026).