🦸🏼‍♂️ How to Get Creative Superpowers (Without Selling Your Soul)


Prefer to read (or listen to) this newsletter in your browser? Click here.

đź’ˇ The Big Idea: AI can give you creative superpowers. But you have to use it intentionally.

There are two camps when it comes to using AI for creative work, and they can’t stand each other.

On one side, the purists refuse to let AI touch their writing. They’re fine using AI for mundane tasks, but the creative process? That’s sacred. To the purists, using AI in the creative process at all is cheating and discredits the work as not really yours if the robot had a hand in making it.

On the other end of the spectrum, you have the junkies. These are the people who wire AI into everything. They couldn’t care less about the process as long as the task gets finished. Push button, done. They’re shipping more than they ever have, and they don’t care how.

To be honest, I can see both sides. But I don’t agree with either one.

The purists are right that something real gets lost when you hand the whole thing over. The junkies are right that refusing to touch AI is a great way to get lapped.

But AI isn’t a switch. It’s a dial. And your best creative work lies somewhere in the middle.

So how do you find that sweet spot? By understanding the creative process and applying AI strategically.

A System for Creativity

The Creativity Flywheel is a simple framework that explains how the creative process works. There are 5 stages in the Creativity Flywheel:

  1. Capture what resonates
  2. Curate what’s useful
  3. Connect your notes and give them context
  4. Cultivate your ideas and let them develop
  5. Create something new from the component pieces

It’s a flywheel because each stage feeds the next. What you capture becomes raw material to curate. What you curate gets connected to what you already know, given context, and becomes the seed of what you create and put out into the world.

When you hit the last step of the flywheel, a funny thing happens: the act of creating something new helps you sharpen your thinking and primes the pump for what to capture next.

Each stage of the flywheel is a different kind of work. Which means AI gets deployed differently at each one.

But used skillfully and intentionally, it amplifies your voice and makes you a more effective creative tastemaker.

Let’s walk through the wheel and find the right setting for each one.

Capture: AI clears the friction, you supply the resonance

Capture is where the Flywheel starts, and it’s the stage AI matters least.

Why? Because AI can't tell you what's worth paying attention to.

This may be surprising to some, but my capture rule has always been to capture what resonates. Not what you think might be important someday. Not what an algorithm decided was relevant. What actually piques your curiosity in this moment.

I grew up playing the violin, which makes sound by drawing a bow across the strings and causing a reverberation through the body of the instrument itself. When you play the violin, you can feel the reverberation. That feeling is very similar to when you come across a potentially useful idea. It hits differently, and it causes a reverberation in your soul.

It’s not something that you feel you might need someday. It’s something that piques your curiosity right now. Those are the types of things you want to capture.

So capture fast, but only capture what resonates.

Goal: Capture what resonates
​Where AI Fits: Eliminating friction when capturing what resonates
​Trap to Avoid: Letting AI decide what’s worth capturing (without the resonance, it's just a bunch of noise)

Curate: don’t delegate the taste

You can’t judge the quality of an idea the moment you have it.

In my experience, we tend to overvalue our ideas when we have them. But if you insert some space between capturing those ideas and adding them to your notes app, you can better see them for what they really are.

Think of yourself as the curator in the museum of your mind. What makes a museum collection valuable is not just what’s there, but also what’s not included. The things that didn’t make the cut end up increasing the value of the ones that do.

So about once per week (sometimes longer), I’ll go through all the things I’ve captured. If I think something is worth keeping, I’ll move it into Obsidian.

Everything else gets deleted.

At this point in the process, you want to add intentional friction. You actually don’t want an effortless way to send things from your capture bucket straight to your notes app. By adding some time and an additional step, you ensure that only the best stuff makes the cut.

I estimate that I only bring over about 10% of the things I capture. The other 90% I discard, trusting that if it’s really important, it’s going to come back to me (it always does).

By its very nature, AI tends toward the average. That’s why it’s great at summarizing.

But curating is not summarizing. Summarizing compresses toward the representative: the gist, the main points, the average of what’s there. That’s exactly what a language model is built to do, and it does it well.

Curating does the reverse. It’s the hunt for the non-average. Curating is deciding what stays and what goes.

That’s taste. And you can’t outsource taste.

Goal: Curate what’s useful
​Where AI Fits: Very little — at most, resurfacing old captures for you to re-judge
​Trap to Avoid: Using AI to summarize or decide what to keep (it averages; curation hunts the non-average)

Connect: where AI actually earns its keep

In his classic book How to Read a Book, Mortimer Adler talks about the 4 levels of reading:

  1. Elementary Reading – where you consider, “What does this sentence say?”
  2. Inspectional Reading – where you consider, “What is this book about?”
  3. Analytical Reading – where you ask organized questions to grok the author’s arguments
  4. Syntopical Reading – where you consider the author’s arguments in context of the other books you’ve read

No one book contains the total answer you’re looking for. And no one idea is the final word. I believe that ideas (and notes) gain value exponentially when they are connected and given a broader context.

IMHO this is where AI really shines – if you’ve captured and curated effectively.

Connecting your ideas is about linking what you’re taking in to what you already know. This is where the best creative thinking actually happens, because creativity is just connecting existing dots in new and interesting ways.

The trouble is that your best dots tend to be buried in notes you forgot you wrote.

So this is the one part of the process where AI does something you genuinely can’t do on your own. It can read across your whole vault and surface connections you’d never spot (i.e., a note from three years ago that speaks directly to the thing you captured this morning).

But an important caveat: AI can only propose connections. It can’t tell you which ones matter, and it’ll happily invent links that sound profound and turn out hollow. So treat AI’s suggestions merely as recommendations.

Connections are most valuable when they are limited and intentional. So don’t just blindly accept whatever connections AI happens to find. Decide for yourself what’s actually worth connecting, and you’ll get a lot more out of the notes and ideas you collect.

Goal: Connect your notes and give them context
​Where AI Fits: Surfacing non-obvious connections across your whole vault
​Trap to Avoid: Blindly accepting the links AI proposes (leads to links that don't mean anything)

Cultivate: AI argues, you decide

To get the most out of your ideas, you need to give them the right environment to grow. Like a seed, you need to give it time to develop in order to see what it really is. You can give it the right conditions to flourish, but you can’t force it before it’s ready.

Think of your PKM system as a greenhouse for your ideas. Once you plant them, you need to let them mature before you start to see the fruit.

One of the worst things you can do for an idea is try to make something out of it before it’s had time to ripen. But when you have a system for developing your ideas, then you don’t feel pressure to force an idea before it’s ready. You know you’ll have more to choose from because the flywheel continues to turn.

This is a great place to use AI as a thinking partner. Pressure-test a half-formed argument. Play devil’s advocate. Ask it to tell you what you’re missing or show you the strongest version of the other side. It’s like having a sharp collaborative partner on call who never gets tired of you asking it questions.

What I won’t let AI do is supply the perspective. You can’t delegate the understanding. The whole point of cultivating an idea is that you’re the one developing a point of view — wrestling it into something you actually believe with conviction.

Use it to sharpen what you think. Not to think for you.

Goal: Cultivate your ideas and let them develop
​Where AI Fits: Thinking partner (pressure-testing, playing devil’s advocate, spotting blind spots)
​Trap to Avoid: Letting AI supply the perspective (you can’t delegate the understanding)

Create: AI can help you get in the reps you need to be great

The last step of the Creativity Flywheel (and the one a lot of PKM folks miss) is to create something new from the component pieces.

I like to think of ideas as individual mental Lego blocks. The pieces may not look all that unique or special, but when you combine them together, you end up creating something new and original.

For a long time, I thought I just wasn’t creative. Then I read Steal Like an Artist by Austin Kleon, and I realized that when you create something new, you’re simply connecting pre-existing dots in new and interesting ways.

The truth is, everything is a remix. The trick is to have good pieces to work with.

Once you have the pieces, though, you still have to make something out of them.

This is where a lot of people get stuck. Writing is hard, and if you’re a perfectionist, looking at a crappy first draft can be painful.

Which is why I actually think asking AI to write for you can ultimately result in better writing.

Your mind works a lot like a water wheel — information must flow in (the Capture phase), and information must flow out (the Create phase). If there’s no outflow, the (fly)wheel stops turning.

Once the wheel gets turning, it starts to build momentum. But getting going can be tough, so when I feel stuck, I have no qualms with asking AI for help.

Sometimes I’ll ask it to help me develop an outline. If I’m feeling really stuck, I may even ask it to write a crappy first draft.

The most important thing is that the flywheel keeps turning.

The line I won’t cross is voice. The entire point of the Flywheel is to publish work that’s unmistakably yours – and the second AI writes it, it loses something. It goes flat. It starts to sound like everything else.

Your voice gets muffled. The taste is gone.

So you still need to go through it, line by line, and own every word. AI can help you get started, but you still have to put in the work to pull it all together.

When you use AI this way, it makes the writing better, but it doesn’t make it faster.

Goal: Create something new from the component pieces
​Where AI Fits: Breaking blank-page resistance (outlines and rough first drafts to keep the flywheel turning)
​Trap to Avoid: Letting AI write it whole and muffle your voice

The Bottom Line: Get help, but keep the part that’s yours

So should you use AI in your creative process? Yes. All of it. Capture, curate, connect, cultivate, create — there’s a place for AI at every stage, and refusing to touch it just means doing hard things the hard way for no reason.

If you want to play the creative game on Master Mode, go for it. But for the rest of us, I hope I’ve shown that AI, used intentionally, can give you creative superpowers.

But the same through line appears in every stage: AI can handle the work around your thinking. It can’t do the thinking. It can carry what resonates, but it can’t feel it. It can propose connections, but it can’t tell you which ones matter. It can draft the words, but it can’t recreate your authentic voice.

So go ahead and use AI. Use it everywhere. It has real, practical value in every part of the creative process.

But remember, it can’t provide the taste.

That part is, and always will be, yours.

— Mike

Practical PKM

A weekly newsletter where I help people apply values-based productivity principles and systems for personal growth, primarily using Obsidian. Subscribe if you want to make more of your notes and ideas.

Read more from Practical PKM

Prefer to read (or listen to) this newsletter in your browser? Click here. 💡 The Big Idea: You Don’t Have to Wait for Someone Else to Fix It Anymore A few weeks ago, an Obsidian update quietly broke one of my favorite plugins. A small tweak to the app destroyed every custom callout in my vault overnight. As frustrating as this was, it’s not the first time something like this has happened. Right before I released LifeHQ, a plugin called Query Control was broken in the exact same way. Last...

Prefer to read this newsletter in your browser? Click here. đź’ˇ The Big Idea: What James Clear told me about how he views using AI in the writing process Last week, I was in Boise for the Craft + Commerce conference. I always love going to Craft + Commerce, but this year was extra special because I got an invite to a private mastermind that happened the day before the conference started. Part of that mastermind was a 90-minute Q&A with James Clear, author of Atomic Habits. James Clear, author...

Prefer to read (or listen to) this newsletter in your browser? Click here. 💡 The Big Idea: The better the tools get, the more the only thing that matters is what you bring to them. There’s a section near the start of Four Thousand Weeks that I can’t stop thinking about. Oliver Burkeman opens the book with a bold prediction made by the famous economist John Maynard Keynes. Back in 1930, Keynes stood in front of a room and told everyone that technology was about to make us roughly eight times...