When you talk to AI, especially in a back-and-forth conversation, something fundamentally different happens than with traditional brainstorming.
You can start with a rough, half-formed idea — something you barely know how to articulate — and the AI doesn’t just respond to the words you say. It listens for what you’re actually getting at. It reflects that back, expands it, organizes it, and often pushes the idea further than you would have on your own in the same amount of time.
In just a few minutes, you don’t walk away with “an idea.”
You walk away with structure.
Sometimes that structure looks like a framework or a strategy. Other times it’s an article outline, or even an entire piece of writing that already feels coherent. The thinking has advanced, not just been captured.
That’s powerful — but it raises an obvious question.
Now What?
If you don’t do anything intentional, those expanded ideas stay exactly where they were born.
They sit buried in chat history, scattered across conversations, effectively lost to time. Which is ironic, because the hardest part — developing the idea — has already happened.
Creativity isn’t the bottleneck anymore.
Capture is.
AI has made idea generation cheap and fast. What it hasn’t solved — at least not yet — is where those ideas go afterward.
My Solution: Turn Conversations Into Artifacts
Here’s what I started doing.
Whenever I have a meaningful exchange with ChatGPT — especially one where an idea becomes clearer, sharper, or more developed — I ask it to turn that thinking into an article or blog post.
Not because it’s finished.
Not because it’s ready for publication.
And not because anyone else needs to see it.
I do it because I want a durable artifact of that thinking. Something that exists outside the conversation. Something I can come back to later.
Why a Blog Post?
Because blog posts force ideas into a shape that’s easy to re-enter.
They’re readable. They’re structured. They’re scannable in a way raw chat transcripts simply aren’t. A developed idea in article form is much easier to revisit than scrolling through a conversation and trying to remember what mattered.
Sometimes, when I come back to one of these posts, I read the whole thing. More often, I skim it. But either way, it does exactly what I want it to do:
It puts me back into the mental state I was in when the idea emerged.
That’s where new ideas tend to come from.
Read Eric’s Mind: My External Idea Shelf
To store these artifacts, I created a separate WordPress site called Read Eric’s Mind.
It’s public-facing, but it’s not something I actively promote or optimize. You’re welcome to look at it if you’re curious, but the purpose isn’t publishing.
It’s retrieval.
That site functions as an external memory — a shelf of semi-developed ideas, a place where thinking doesn’t disappear just because I moved on to something else.
Some of the posts are rough. Some are more polished. None of them are there because they’re “done.”
They’re there because they’re worth remembering.
How I Keep Ideas From Turning Into a Junk Drawer
Capturing ideas is only half the solution. If everything goes into one giant pile, all you’ve done is create a more sophisticated way to forget things.
So I do one simple thing that makes all the difference: I separate categories from tags.
When an idea gets turned into an article, it goes into a broad category based on what it’s fundamentally about. Book ideas live in a book category. AI-related thinking goes into an AI category. Business or product concepts go into their own bucket. Personal or lifestyle ideas go somewhere separate.
Those categories answer a big question:
What area of my life or work does this belong to?
Then I layer on tags.
Tags are more specific. They capture themes like time management, productivity, idea capture, systems, writing, strategy, or personal organization. An idea lives in one category, but it can carry many tags.
That lets me browse broadly when I’m in a certain headspace, or search narrowly when I’m chasing a specific thread. There’s a difference between asking, “What have I ever thought about AI?” and “What have I said about AI and time management?” Both questions matter.
Over time, patterns emerge — not because I planned them, but because ideas that share categories and tags start to find each other.
One Drive, Many Ideas
What surprised me most is how quickly this compounds.
On a single 15-minute drive, I might explore one idea, branch into a sub-idea, realize there’s a related angle, and end up with multiple articles worth of thinking.
Without a capture system, most of that value would evaporate.
Instead, it accumulates.
Why This Matters Long-Term
Over months and years, this creates an interesting effect. I can look back and recognize that an idea existed long before it had a name. That something wasn’t random — it was forming.
That perspective is incredibly valuable when you’re building anything long-term.
Ideas rarely arrive fully finished. They mature.
But only if you don’t lose them.
The Bigger Point
AI makes idea generation cheap. Conversation makes idea development fast.
But idea capture is what makes creativity compound.
If you’re using AI regularly and not thinking about where those ideas live afterward, you’re leaving value on the table. You don’t need a perfect system — just a place where ideas don’t disappear.
For me, that place is a simple website. For you, it might be something else.
But without some form of capture, the gold stays in the ground.