Vibe Learning — Issue #20
The AI Learning Flywheel
https://vibelearning.ca/index.html
How to Build a Learning System That Gets Better Every Time You Use It
We started this journey by asking a simple question:
What does learning look like when AI changes the rules?
Across this series, we’ve explored curiosity.
Critical thinking.
Feedback.
Deliberate practice.
Second brains.
Learning in public.
Continuous reinvention.
But these ideas aren’t separate.
They connect.
And when you connect them, something powerful happens.
You stop treating learning as an activity.
You start treating it as a system.
A system that can become better every time you use it.
Welcome to the AI Learning Flywheel.
Stop Thinking About Learning as a Straight Line
Most education follows a familiar path:
Learn → Test → Finish
Take the course.
Pass the exam.
Get the certificate.
Move on.
But real learning doesn’t work like that.
You learn something.
You try it.
You fail.
You discover something unexpected.
You ask another question.
You improve.
You explain it to someone else.
That explanation exposes another gap.
So you learn again.
Learning is not a line.
It’s a loop.
And AI can dramatically accelerate that loop.
The Vibe Learning Flywheel
Here’s the complete system:
Curiosity → Explore → Understand → Practice → Feedback → Reflect → Connect → Create → Share → Reinvent → Curiosity
Each stage feeds the next.
And every trip around the loop makes the learner stronger.
Let’s break it down.
1. Curiosity — Start With Questions
Every learning journey begins with curiosity.
Not:
“What do I need to memorize?”
But:
“What do I want to understand?”
Curiosity creates momentum.
AI makes curiosity almost frictionless because every question can immediately lead somewhere.
Try:
“I’m interested in [topic]. Give me five surprising questions about it that would lead me beyond the obvious beginner-level information.”
Don’t start by asking AI for everything you need to know.
Start by discovering what is worth wondering about.
2. Explore — Map the Territory
Before going deep, understand the landscape.
Suppose you want to learn machine learning.
You could immediately watch a 10-hour course.
Or first ask:
“Create a map of the major concepts involved in machine learning. Show me how they connect and which concepts I should understand first.”
Now you have context.
You know where you’re going.
AI is extremely useful here because it can help turn an unfamiliar subject into a navigable map.
But remember:
The map isn’t the journey.
Eventually, you have to start walking.
3. Understand — Don’t Settle for the First Explanation
AI can explain the same concept dozens of ways.
Use that.
If an explanation doesn’t click, don’t assume you’re incapable of understanding it.
Change the explanation.
Try:
“Explain this using a real-world analogy.”
Then:
“Now explain it without the analogy.”
Then:
“Give me an example.”
Then:
“Give me a counterexample.”
Then:
“Ask me to explain it back to you.”
The goal isn’t receiving an explanation.
The goal is reaching the point where you can produce one yourself.
That’s an important distinction.
4. Practice — Turn Information Into Ability
This is where learning becomes real.
You can read about programming for months without becoming a programmer.
You can watch videos about public speaking without becoming a speaker.
You can study writing without becoming a better writer.
Knowledge must eventually become action.
Ask AI:
“Give me a practical challenge that requires me to use what I’ve just learned. Don’t give me the solution unless I ask for help.”
Now AI becomes something more valuable than an answer machine.
It becomes a practice environment.
5. Feedback — Shorten the Improvement Cycle
Traditionally, feedback could take days.
Submit something.
Wait.
Receive comments.
Try again.
AI can reduce that cycle to minutes.
Create something.
Get feedback.
Improve it.
Repeat.
But don’t ask:
“Is this good?”
That’s too easy.
Ask better questions:
“What is the weakest part of this?”
“What would an expert improve first?”
“What mistake am I repeating?”
“Give me one improvement to make before I continue.”
Feedback is most useful when it leads directly to another attempt.
6. Reflect — Turn Experience Into Learning
AI makes it incredibly easy to keep moving.
That’s also a problem.
Sometimes we need to stop.
Reflection asks:
What actually happened?
What did I understand?
What confused me?
What mistake taught me the most?
What should I do differently next time?
Try ending a learning session with:
“Ask me five reflection questions about what I just practiced. Don’t answer them for me.”
This is important.
AI shouldn’t do your reflection.
It should provoke it.
7. Connect — Build Your Second Brain
Now connect today’s learning with what you already know.
This is where your second brain becomes valuable.
Instead of storing isolated facts, look for relationships.
Ask:
“Here are five ideas I’ve been exploring. What connections, patterns, or tensions exist between them?”
This can reveal something far more valuable than another summary.
It can create new insight.
As we explored in Issue #17, the real opportunity isn’t simply having AI organize existing thoughts.
It’s using AI to help synthesize diverse information into connections you might not have discovered yourself.
Chaos becomes clarity.
And sometimes clarity becomes a completely new idea.
8. Create — Make Something That Didn’t Exist Before
Eventually, learning needs an output.
Build something.
Write something.
Teach something.
Solve something.
Design something.
Creation forces multiple layers of understanding to work together.
If you’re learning AI, build a small AI workflow.
If you’re learning JavaScript, build an application.
If you’re learning marketing, create a campaign.
If you’re learning history, build an argument around competing interpretations.
If you’re learning leadership, apply one principle to a real situation.
The question changes from:
“Do I understand this?”
to:
“What can I do with this?”
That’s a much stronger test.
9. Share — Make Your Thinking Visible
This was the focus of Issue #18.
Learning in public creates another feedback loop.
You explain what you’ve learned.
Someone asks a question.
You realize your explanation isn’t complete.
Someone disagrees.
You investigate.
Someone adds an example you hadn’t considered.
Suddenly, publishing isn’t just communication.
It’s part of learning.
Try the simple rule:
Learn one thing. Teach one thing.
Your audience doesn’t need you to know everything.
They need you to communicate something useful clearly.
10. Reinvent — Use Learning to Change What You Can Do
This was the central idea of Issue #19.
Learning should eventually change your capabilities.
Ask:
“What can I do now that I couldn’t do before?”
That’s a powerful measure of progress.
Maybe you can:
• automate a task
• build an application
• understand financial data
• communicate more clearly
• teach a difficult concept
• create better presentations
• analyze research
• use a completely new tool
Learning becomes transformation.
And transformation creates something interesting.
New curiosity.
Which sends you around the flywheel again.
Why It’s a Flywheel
A flywheel is difficult to move at first.
You push.
It moves slightly.
You push again.
It moves faster.
Eventually momentum begins helping you.
Learning works the same way.
Your first programming project is difficult.
Your tenth is easier.
Your first article takes hours.
Eventually ideas flow faster.
Your first AI conversations may be basic.
Over time you learn how to question, challenge, refine, and explore.
Each learning cycle improves the next one.
That’s the flywheel effect.
AI Should Accelerate the Loop — Not Remove It
This is perhaps the most important lesson from the entire series.
AI can remove friction.
That’s useful.
But not all friction is bad.
Struggling with a problem can build understanding.
Trying to remember strengthens retrieval.
Writing an explanation forces clarity.
Making mistakes creates feedback.
AI becomes harmful to learning when it removes the cognitive work we actually needed to perform.
The goal isn’t:
AI does the learning for me.
The goal is:
AI helps me become a better learner.
Know When NOT to Use AI
Sometimes the best AI prompt is no prompt.
Try solving the problem first.
Write your first paragraph yourself.
Brainstorm five ideas before requesting twenty more.
Explain the concept from memory.
Make a prediction before asking AI.
Then compare.
This gives AI something incredibly valuable to work with:
your thinking.
Without that, AI can easily become a substitute for thought instead of a partner in it.
Build Your Personal AI Learning Team
You don’t have to use AI in the same role every time.
Give it different jobs.
The Tutor
Explains difficult concepts.
The Socratic Teacher
Asks questions instead of providing answers.
The Challenger
Looks for weaknesses in your reasoning.
The Practice Coach
Creates progressively harder exercises.
The Reviewer
Critiques your work.
The Connector
Looks for relationships across different ideas.
The Reflection Coach
Asks questions about your learning process.
The Creative Partner
Helps transform knowledge into something new.
The important word is role.
Tell AI what kind of thinking you want it to support.
A Master Prompt for Vibe Learning
Try this when starting something new:
“Act as my learning coach for [TOPIC].
Your goal is not to give me answers as quickly as possible. Your goal is to help me develop real understanding and independent ability.
Start by discovering what I already know.
Help me identify the most important concepts.
Teach one concept at a time.
Ask questions to test my understanding.
Give me practical exercises.
Let me attempt them before providing solutions.
Give specific feedback on my attempts.
Increase the difficulty as I improve.
Periodically ask me to explain concepts in my own words.
Help me connect new ideas with things I’ve already learned.
At the end of each session, ask me to reflect on what changed in my understanding and recommend the best next challenge.”
Notice what this prompt does.
It doesn’t ask AI to teach you everything.
It asks AI to create conditions for learning.
The 30-Minute Vibe Learning Session
You don’t need hours every day.
Try this:
5 Minutes — Explore
Ask one interesting question.
5 Minutes — Understand
Study one concept deeply.
10 Minutes — Practice
Use the concept without asking AI to do the work.
5 Minutes — Feedback
Ask AI to review your attempt.
5 Minutes — Reflect
Write down:
What did I learn?
What mistake did I make?
What’s my next question?
That’s thirty minutes.
Repeated consistently, it becomes a powerful learning habit.
The Weekly Vibe Learning Review
At the end of every week, review your learning.
Ask yourself:
What did I learn?
What did I actually build?
What did I struggle with?
What feedback changed my thinking?
What ideas connected?
What did I share?
What can I now do that I couldn’t do last week?
And finally:
What am I curious about next?
You’ve completed the flywheel.
Now start another rotation.
Don’t Measure Learning by Content Consumed
This is another major shift.
Don’t measure:
10 videos watched.
300 pages read.
15 hours of courses completed.
Measure:
Problems solved.
Ideas explained.
Projects created.
Mistakes corrected.
Questions generated.
Connections discovered.
Skills gained.
Things taught.
Capabilities changed.
Consumption measures activity.
Capability measures learning.
The Ultimate Vibe Learning Challenge
Choose something you genuinely want to learn.
Then spend the next 30 days running it through the flywheel:
Curiosity
↓
Explore
↓
Understand
↓
Practice
↓
Feedback
↓
Reflect
↓
Connect
↓
Create
↓
Share
↓
Reinvent
↓
Repeat
Don’t try to become an expert in 30 days.
Try to become noticeably more capable.
That’s enough.
Because capability compounds.
20 Issues Later…
This series started with a simple premise:
AI is changing learning.
But after 20 issues, I think the bigger lesson is different.
AI isn’t simply changing what we can learn.
It’s changing our relationship with learning itself.
We can have explanations on demand.
Practice on demand.
Feedback on demand.
Questions on demand.
Brainstorming on demand.
Coaching on demand.
The scarcity is no longer information.
The scarcity is:
Curiosity.
Attention.
Effort.
Judgment.
Reflection.
Action.
Those remain ours.
Final Reflection
AI may become the most powerful learning technology we’ve ever created.
But technology alone doesn’t create learners.
Learners do.
The advantage won’t belong to people who ask AI to think for them.
It will belong to people who use AI to:
question more deeply,
practice more deliberately,
connect ideas more creatively,
reflect more honestly,
create more boldly,
and
keep becoming something new.
That is the AI Learning Flywheel.
That is the Infinite Learner.
And that is Vibe Learning.
What’s Next?
Issue #20 completes this chapter of Vibe Learning.
But learning doesn’t have a finish line.
The next chapter can go deeper into the practical side of AI-powered learning: building personalized AI tutors, creating learning agents, designing knowledge systems, mastering difficult subjects, developing future-proof skill stacks, and experimenting with entirely new ways of learning.
Because the most interesting question isn’t:
“What do you know?”
It’s:
“What are you ready to learn next?”
Question for readers
If you could use the AI Learning Flywheel to become noticeably better at one thing over the next 30 days, what would you choose?