Learn Anything Faster: How to Use the Feynman Technique with AI

You've probably heard this before: "Just explain it like you're teaching a five-year-old."

It sounds simple. It sounds like it works. And then you try it and realize you have no idea what you're actually talking about.

That's the Feynman Technique. Also very misunderstood.

People read something, try to explain it back, feel confused, and assume they're just not smart enough. What they're actually doing is exposing a gap in their understanding-which is exactly the point. But without the right structure, without a way to actually fill that gap, you just end up frustrated.

Here's where AI fits.

Feynman Technique isn't magic. It's a thinking tool. And when you pair it with AI-not as a replacement for your brain, but as a sparring partner-you can compress months of learning into weeks. Not because AI does the work for you, but because it forces you to think harder, faster, and more clearly.

What Feynman Technique Is

Richard Feynman was a physicist who won the Nobel Prize. That bloke had a reputation for understanding things deeply.

His method was simple:

1. Pick a concept - Choose something you want to understand
2. Teach it - Explain it in simple language, as if you're teaching someone with zero background
3. Identify gaps - When you struggle to explain something, you've found what you don't actually know
4. Review and simplify - Go back to the source material, relearn, and try explaining again in even simpler terms

The magic isn't in the explaining. It's in the struggle.

When you can't explain something clearly, your brain is telling you something. You don't actually understand it yet. You've memorized it. You've skimmed it. But you haven't integrated it.

Most people stop there. They feel bad, they give up, or they move on to the next thing.

That's where 99% of learning fails.

The Feynman Technique only works if you actually do something with that gap. And that's where AI becomes your secret weapon.

Why AI + Feynman Hits Different

Here's the honest version: AI is not a substitute for thinking. It's a tool that makes thinking faster.

When you use AI as a crutch-asking it to explain something so you don't have to struggle-you're wasting it. You're also wasting your time. You'll forget it by tomorrow.

But when you use AI as a thinking partner, something shifts.

You explain your understanding to AI. It pokes holes in your logic. You refine. You go deeper. You come back with a better explanation. It challenges you again. Repeat.

This is the Protege Effect in action. The teacher learns more than the student. When you're forced to articulate your thinking, you learn faster than if you just consume information.

AI accelerates this cycle. It's patient. It doesn't judge. It asks follow-up questions. It catches inconsistencies you'd miss on your own.

And crucially: it remembers what you said, so you can build on it over multiple conversations.

How to Use This

Let's say you're learning something real. Programming. Data science. Guitar. Copywriting. Doesn't matter.

Step 1: Learn the basics (but not too much)

Don't watch 10 tutorials. Don't read the whole book.

Get 20% of the way there. Enough to have something to explain. Enough to know you're confused about the rest.

Use AI to find the best resources fast:

I want to learn [topic]. Give me:
- The 3 best beginner resources for my level
- A summary of each
- Why each one is helpful
- What I should focus on first

Pick one. Spend 30 minutes. That's it.

Step 2: Attempt to explain it (don't worry about being wrong)

Open a new chat with your AI. Don't ask it to explain the topic. You explain it.

I'm learning about [topic]. Here's my current understanding:

[Your explanation - be specific, include examples, say what you think happens]

What am I getting right? What am I missing? Where's my logic breaking down?

This is uncomfortable. Good. That discomfort is the signal that learning is happening.

Step 3: Let AI expose your gaps

AI will point out inconsistencies. It will ask clarifying questions. It will show you where your mental model is incomplete.

Don't just read its response and move on. Engage with it.

You said I'm confusing [concept A] with [concept B]. Can you explain the difference 
in a way that makes it obvious why they're not the same?

Step 4: Refine your explanation

Now you go back to your source material. You read more carefully. You look for the specific thing you got wrong.

Then you come back to AI and try again:

Okay, I think I understand now. Let me try explaining [concept] again:

[Your refined explanation]

Is this more accurate? What else am I missing?

Each cycle, your explanation gets tighter. Your understanding gets deeper.

Step 5: Teach it to someone (or write about it)

This is the final test. Can you explain it to a real person? Can you write it down clearly?

I'm going to explain [topic] to someone who knows nothing about it. 
Here's what I'm planning to say:

[Your explanation]

Does this make sense? Will they understand? What should I add or clarify?

If you can do this without AI's help, you've learned it.

Real Examples

Example 1: Learning Python loops

You watch a 10-minute tutorial on for loops. You think you get it.

You open AI and say:

I'm learning Python. Here's what I understand about for loops:

A for loop repeats code a certain number of times. You write "for i in range(5)" 
and it runs the code 5 times. The "i" is like a counter.

AI responds: "Close, but you're missing something important. What do you think i actually represents on each iteration? And what happens if you use i inside the loop?"

You realize you don't actually know. You go back to the tutorial. You see that i is the current value in the sequence, not just a counter. You try again.

Okay, so "for i in range(5)" means i takes the values 0, 1, 2, 3, 4 one at a time. 
Each time through the loop, i is a different number. So if I print(i), I'll see 0, then 1, then 2...

AI: "Exactly. Now, what if you wanted to loop through a list of names instead of numbers? How would that work?"

You struggle. You think. You realize you need to understand how for loops work with different data types. You go back and learn that. You come back with a better explanation.

By the end, you don't just know what a for loop is. You understand why it works the way it does. You can explain it. You can use it in different contexts.

That's learning.

Example 2: Understanding copywriting principles

You read about the "pain-agitate-promise" framework. You think it's just a formula.

You tell AI:

I'm learning copywriting. Here's what I think "pain-agitate-promise" means:

You identify a problem the customer has (pain). You make them feel bad about it (agitate). 
Then you tell them your product solves it (promise).

AI: "You've got the structure, but you're missing the why. Why would you want to make someone feel bad? Isn't that counterintuitive?"

You realize you don't actually understand the psychology behind it. You go back and learn about emotional triggers, urgency, and why people buy things. You come back:

Oh, I see. You're not making them feel bad to be mean. You're making them *aware* of the problem 
so they're motivated to solve it. If they don't feel the pain, they won't care about your solution. 
The agitate step is about making the problem real and urgent.

AI: "Now you've got it. Can you give me an example of this in action?"

You write a real example. AI critiques it. You refine it. You've now learned not just the framework, but how to apply it.

Prompts That Work

Use them.

For initial explanation:

I'm learning about [topic]. Here's my current understanding:

[Your explanation - be as specific as possible]

What am I getting right? What am I missing? Where's my logic breaking down?

When you're stuck:

I'm confused about [specific thing]. I think it means [your guess], but I'm not sure. 
Can you explain why I'm right or wrong? What's the actual difference?

For deeper understanding:

Explain [concept] from three different angles:
1. How it works mechanically
2. Why it matters (what problem does it solve?)
3. How it connects to [related concept]

Before you teach it:

I'm going to explain [topic] to someone who knows nothing about it. 
Here's my explanation:

[Your explanation]

Will they understand? What should I add, remove, or clarify?

To catch yourself fooling yourself:

I just explained [topic] to you. Now, can you ask me a question that would reveal 
if I actually understand it or if I'm just repeating what I memorized?

Feeling Like You're Learning When You're Not

Here's the dangerous part.

AI is so good at explaining things that you can feel like you understand without actually understanding.

You read AI's explanation. It makes sense. You feel smart. You move on.

Two days later, you can't remember any of it.

That's not learning. That's consumption.

Real learning feels different. It feels like struggle. It feels like you're building something. It feels like you're connecting dots.

If you're just reading AI's responses and nodding along, you're wasting your time.

Feynman Technique only works if you're the one doing the explaining. AI is just the mirror that shows you where your explanation breaks down.

Notice if you're doing this:

• You ask AI a question and read the answer without trying to explain first
• You feel good after reading AI's explanation but can't explain it back
• You're collecting information instead of building understanding
• You're moving fast but not retaining anything

If any of that is true, you're using AI wrong.

How to Know You've Actually Learned It

You've learned something when:

1. You can explain it without AI - You don't need to look anything up. You can articulate it clearly.

2. You can apply it to new situations - You understand the principle well enough to use it in contexts you haven't seen before.

3. You can teach it to someone else - Not just recite it, but actually help someone else understand it.

4. You remember it a week later - You don't need to relearn it from scratch.

5. You can defend it - Someone challenges your understanding and you can explain why you think you're right.

If you can do all five of those things, you've learned it.

If you can only do one or two, you haven't yet. Go back to step 2.

The Secret

The Feynman Technique isn't about explaining things simply. It's about exposing what you don't know.

AI doesn't replace that exposure. It accelerates it.

Instead of struggling alone for hours, you struggle with a partner who asks better questions. Instead of guessing whether you understand, you get immediate feedback. Instead of moving on to the next thing, you actually integrate what you just learned.

This is why people who use this method learn 10x faster than people who just consume content.

Not because they're smarter. Not because they have more time.

Because they're actually learning instead of just scrolling.

The difference is the struggle. The difference is the feedback loop. The difference is forcing yourself to articulate your thinking instead of passively absorbing information.

AI makes that loop tighter and faster.

But only if you use it right.

---

Here's what to do next:

Pick one thing you're trying to learn right now. Something specific. Not "I want to learn programming"-something like "I want to understand how APIs work" or "I want to learn the basics of video editing."

Open your AI. Don't ask it to explain it. Explain it yourself first. See where you get stuck.

That's where the learning actually begins.

AI can help. For real! AI can definitely improve your mental health or help you learn way faster, but you need to be the one putting in effort.