I had 1,400 notes and could find none of them.
Eight years of clipped articles, half-thoughts, meeting scraps, and "I'll definitely use this later" ideas — all dumped into a notes app that had quietly become a graveyard. Writing things down felt productive. Finding them later was impossible. My "second brain" had amnesia.
Then I rebuilt the whole thing around AI, and for the first time my notes started giving things back. Here's the system.
A traditional note app stores; an AI second brain retrieves and connects. The shift is from organizing notes by folders and tags (which you'll never maintain) to dumping everything in messily and letting AI find, summarize, and link ideas on demand. You stop being the librarian. You just ask, "what did I write about X," and the connections you forgot surface themselves. Capture messy, retrieve smart.
Here's the lie of traditional note-taking: that capturing an idea is the valuable part.
It isn't. Capturing is trivial. Retrieval is everything — and it's the part every folders-and-tags system fails at. I had brilliant past-me notes I never saw again because I couldn't remember the right keyword, the right folder, or that I'd even written them.
A note you can't find is worse than no note, because you think you have the knowledge. The whole premise of a second brain — that your past thinking compounds — collapses if you can't reach the past thinking. AI fixes exactly this part. It's the same lesson running through the honest truth about AI productivity tools: the leverage shows up when AI takes over a whole job you were quietly failing at — here, recall — rather than speeding up one you were already doing. The Nielsen Norman Group has long noted how much knowledge work is lost not to bad capture but to bad retrieval, which is precisely the gap this closes.
The core difference is who does the organizing. In the old model, you organize, constantly, forever — and you fall behind in a week.
In an AI second brain, you barely organize at all. You capture messily, and the intelligence lives in retrieval. You ask questions in plain language and the system finds, summarizes, and connects across everything you've ever saved.
The mental shift:
That single inversion is what made eight years of notes suddenly useful. I stopped maintaining a filing system I always abandoned, and started just asking.
The day-to-day is almost embarrassingly simple.
Capture is brainless. Idea, link, meeting scrap, voice memo — it all goes into one place with zero sorting. No folder decision, no tag agonizing. The friction that used to stop me from capturing is gone, so I capture more, which makes the brain richer.
Retrieval is a conversation. When I need something, I ask: "what have I written about pricing?" or "summarize everything I've saved on this client." The AI pulls from across years of notes and hands me a synthesis — not just a list of files, an actual answer.
Stop organizing notes on the way in. Capture them messy and let AI organize them on the way out.
Connections surface on their own. This is the part that genuinely surprised me. The AI links ideas I never knew were related — a note from two years ago suddenly relevant to a problem today. My old app could never have shown me that. I'd forgotten the note existed. It's the same shift from doing the work to directing it that I described in the quiet way AI agents run my side business: you stop being the operator and start being the one who just asks.
A few weeks in, I was stuck on a hard decision and, half out of habit, asked my notes about it.
The AI surfaced three things: a framework I'd saved long ago and forgotten, a related decision I'd made before and the outcome, and a relevant quote from an article I'd clipped. In ten seconds it handed me my own accumulated thinking on the exact problem in front of me.
That's when it clicked. A second brain isn't a place to store thoughts. It's a place that hands them back at the moment you need them. My old note app stored. This one remembers, and reminds. That's the entire difference, and it's enormous.
| Old note app | AI second brain | |
|---|---|---|
| Organizing | manual, constant | mostly automatic |
| Finding things | keyword guessing | ask in plain language |
| Connections | you make them | surfaced for you |
| Capture friction | high (where does this go?) | near zero |
| Notes that resurface | rarely | constantly |
The right column is just less work and more value, which is the rare combination that makes a system actually stick.
The hardest part of switching wasn't technical. It was psychological. I had to unlearn a decade of beliefs about what "good note-taking" meant.
For years I'd absorbed the gospel of elaborate systems — nested folders, careful tags, linking everything by hand, color codes, the whole productivity-influencer apparatus. And I'd genuinely believed that the act of organizing was where the value lived. That a beautifully structured note was a better note.
It wasn't. It was just more expensive. All that organizing was work I did for the system rather than work the system did for me. And because it was effortful, I always fell behind, felt guilty, and eventually abandoned it — which is the real reason all my previous note apps became graveyards. They demanded maintenance I couldn't sustain, and a system you can't sustain is a system you don't have.
The AI second brain forced me to accept something that initially felt like cheating: messy is fine. A pile of unsorted, untagged, half-formed notes is more valuable in this model than a pristine, lovingly-organized one, because the messy pile is bigger, captured faster, and the retrieval layer doesn't care about your folder structure. I stopped grooming my notes and started just feeding them. My note volume tripled almost immediately, simply because the friction of "where does this go" disappeared.
There's a small grief in letting go of the beautiful system. It felt productive. It looked impressive. But it was a hobby disguised as a workflow. What I actually wanted was never a tidy library — it was my past thinking, handed back to me at the right moment. Once I admitted that, the tidy library stopped being the goal and started being the thing in the way.
Letting go of organizing doesn't mean letting go of everything. There's exactly one discipline I kept, and it's the one that makes the whole messy system work: capture in your own words.
When I save an idea, I add a sentence of why it mattered to me — not just the clipped article, but "this reframes how I think about pricing" or "use this for the onboarding problem." That tiny annotation is the hook the retrieval layer grabs onto later. A raw link is findable; a raw link plus your reaction is yours, and it surfaces at the right moment because it's tagged with your actual thinking instead of someone else's headline.
This is the difference between a second brain and a bookmark folder. A bookmark folder stores other people's thoughts. A second brain stores your reactions to other people's thoughts — and reactions are what you actually want back later. The article was never the valuable part. What you thought about the article was.
It costs maybe five seconds per note. It's the only "organizing" I do, and it's not really organizing — it's just thinking out loud at the moment of capture, while the thought is fresh. Everything else, I happily leave to the machine. One small habit, kept; a decade of folder-grooming, finally dropped.
If your own notes app has become a graveyard, try the inversion for a week — capture messily, ask on the way out — and see whether your past thinking finally starts handing itself back.
Q: Do I have to import years of old notes? Helps, but not required. Even starting fresh, the retrieval and connection benefits show up within weeks. If you can import the old stuff, the dormant value in those forgotten notes is the best part.
Q: Isn't messy capture just chaos? It would be, in an old system. The point is that AI handles the order on retrieval, so the chaos never has to be sorted by you. Messy-in is fine when smart-out is guaranteed.
Q: Will it surface wrong or irrelevant connections? Sometimes — it's suggesting, not deciding. But a few off-target links are a tiny price for the genuinely useful ones you'd never have found yourself. You stay the judge of what's relevant.
Q: Is this overkill for someone with few notes? Even a small, growing note collection benefits from frictionless capture and plain-language retrieval. And the earlier you start, the more your thinking compounds over the years.
For eight years I confused writing things down with remembering them. They're not the same. A note you can't retrieve is just a polite way of forgetting.
The AI second brain fixed the part that was actually broken — not capture, but recall. Now I dump ideas in without a thought and pull them back out in conversation, connections and all. My past thinking finally compounds instead of decomposing in a folder.
If your notes app has quietly become a graveyard too, the answer probably isn't a tidier filing system. It's stopping being the librarian and letting AI hand your old ideas back to you, right when you need them.
What would you ask your past self, if you could actually find what they wrote?
One person, output that looks like five. It isn't about working more hours — it's about a kind of leverage teams rarely have.
One idea a week to a published issue in under an hour. The boring system behind a newsletter I never dread sending.
Behind a lot of lean, profitable companies is the same small stack of AI tools. Here's what's actually running the show.
Comments
Sign in to join the conversation
No comments yet. Be the first to share your thoughts!