Advanced · Critical thinking and your own skills · module 127 of 155
Learning with AI without outsourcing the thinking
Use AI as a practice partner that makes you think, instead of as a machine that takes over the thinking.
After this lesson
You use AI to quiz yourself, get explanations and ask for feedback on your own work, while you first think for yourself.
In view · An improvement loop
Learning with AI as a practice partner
Try an answer or solution before you ask for help.
Ask for a hint, a simpler explanation or feedback on your own work.
Have yourself quizzed and come back to what you learned at set times.
Part 1 of 3 · Yourself first
Your preference for less movement is on. Choose each part yourself.
Learning takes your own effort
You remember something better when you make an effort yourself: recalling an answer from memory, trying a problem yourself first, explaining something in your own words. When AI gives the full answer straight away, it feels efficient, but you often learn less from it. You recognise a good solution, but cannot produce it yourself. That applies at school, but also to new skills at work, such as learning to use a spreadsheet or getting to know a new field.
AI as a practice partner
Ask AI to quiz you one question at a time and to give feedback only after your answer (lesson 101). Have a concept explained at a simpler level, with an example from your own field. Ask for a hint instead of the solution. Present your own explanation and ask what is missing or wrong. Check facts in a reliable source, because an explanation can also contain errors (lesson 103). That way you remain the one doing the thinking, and AI serves as a sparring partner.
Being honest about what is yours
Follow the agreed rules on AI help, for example at school or on a training course, and state how you used AI when asked. Having an assignment done for you and submitting it as your own work is not learning and can count as fraud. Keep your own first version alongside what you improved with AI help; that way you see your progress. Plan review at fixed times, such as the review sessions in this course, so that knowledge sticks (lesson 95).
New practice example
From explanation to a usable result.
Learn with AI by thinking for yourself first. Let AI quiz you, give you hints and check your own explanation.
Fictional practice situationFictional: a new employee of a municipal department has to learn the rules for parking permits. A colleague sends her a summary made with AI every day.
View a possible solution
This is how it can be done
A better approach: she first reads the official rules herself, then answers five questions without help and has AI compare her answers with the rules. Where she missed something, she asks for an explanation with an example and checks it in the official text. After a week, AI quizzes her again.
Why this works
She recalls the knowledge herself and uses AI to test her understanding. The official text remains the source.
Make it slightly more difficult
Write a prompt that has AI quiz you on a topic from this course, without giving the answers straight away.
Extra practice, at your own pace. You do not need to submit anything. This does not count towards an automatic assessment.
Your workplace · learning by doing
Make something. See what changes.

Checking the AI connection… You can already view the written example.

Present your assignment to Ende.
Read the answer, change one thing, and compare.
Previously held
New version
You remain in control
What do you want to improve in your next attempt?
On this device only. AI answers remain only in this page memory. This is self-review, not a grade.
Why this approach works
She recalls the knowledge herself and uses AI to test her understanding. The official text remains the source.
A step further
Write a prompt that has AI quiz you on a topic from this course, without giving the answers straight away.
Try & discover · no obligation
Become the source checker yourself
Only mark statements that the source does not support. Then check your selection.
Fictional: five colleagues use AI to learn a new spreadsheet function. Who really learns something new?
Select which sentences you would correct above. The answer remains hidden until you check your selection.
This trial works locally with fictitious data and does not count towards your lesson completion. It starts anew with a new lesson or reload.
A prompt to get you started
I want to learn what the context window of a language model is. Do not give me the explanation straight away. First ask me two questions to see what I already know. Then explain at the level that matches my answers, with an example from office work. Finish with three practice questions, one at a time, and give feedback only after I have answered.
Apply it
A colleague who has to learn a new field has summaries made every day and skims them. After a month she still cannot explain the basics herself.
Which approach best fits this situation?
- Have even more summaries made.
- Have her answer questions herself first and use AI to quiz her and check her own explanations.
- From now on, have AI answer all questions, so she doesn’t need to know it herself.
With JavaScript, you can check your approach and receive immediate feedback.
Test your insight
You want to really learn how to use a new spreadsheet function. Which approach helps most?
- Have AI give you the complete formula and copy it.
- First have a go yourself, then ask for a hint and have your own formula checked.
- Have an explanation made and save it without practising.
Extra practice
Choose a topic you want to learn. First try to answer three questions yourself, then use AI as a practice partner. Note what you already knew and what you learned.
My own answer beforehand: “The context window is the model’s memory.” After the exercise: it is the amount of text the model can take into account for one answer, including instructions, conversation and room for the answer; whatever falls outside it, the model cannot use. I had interpreted memory too broadly. I then answered the three practice questions without help.

