Constraint is not a limiting box.
It’s what forces an idea to take flight.
Without constraints, ideas don’t more lift.
They just get produced.
And right now, it’s never been easier to produce something that looks finished.
AI can generate:
content
campaigns
messaging
in seconds.
I use it all the time at my desk, to explore directions, pressure-test ideas, and refine thinking faster than I could on my own.
But I try to be very intentional about how I use it.
Because the real risk isn’t bad work.
It’s work that looks good enough to move forward… and does nothing.
After all, AI is built to make us feel like our ideas are the best and we’re rock star writers.
So here’s what my workflow actually looks like in practice.
I usually start with a rough idea—something I’ve heard, read, or noticed.
Recently, it was this:
AI takes confusion and creates order.
It can also remove the need for humans to think hard.
Humans, on the other hand, create meaning by working through the mess.
I brought that into a working session and started shaping it by bringing in other material that I thought had a similar thread.
First pass:
→ what are the takeaways?
→ where are the overlapping ideas?
It came back structured, clear, usable.
Which sounded okay, but I didn’t feel like the idea was really getting to the point. The reasoning was repeating the existing content, not creating new connections or insights. I tried probing, and injecting other ideas, but still, something felt off.
So I paused, said this wasn’t what I was after, and set it aside.
No pushing forward. No forcing a clean version just to have something done.
I sat with it instead, with open tabs everywhere (47 at the moment, which is probably not a best practice… but here we are). I let different angles and different threads bounce around for a while, read some additional material.
Then it clicked.
It’s not just about AI versus human thinking.
It’s about when clarity shows up, and what gets skipped when it arrives too early.
It’s about messy thinking having value. It’s about constraint doing its job, keeping the idea in motion long enough to become stronger before it gets polished and pushed out.
That’s when I opened the AI conversation back up.
This time, the prompts changed:
→ write it from this angle
→ bring in this specific example
→ make the outcome clearer
And suddenly, it had weight and I finally had something that could be posted.
But I don’t stop using AI there, either. That’s where it can help massage, tighten, and polish. It can support the work beautifully. It just can’t replace the part where I have to know what I’m actually trying to say.
That pause, that frustration, that “this isn’t it yet” feeling—that’s not inefficiency.
That’s the work.
And with clients, that part doesn’t happen in a tool.
It happens in the room.
I see this with leadership teams all the time. Something gets created quickly. It sounds right. It checks the boxes.
But no one has actually worked through the guts behind it.
No real debate.
No push and pull.
No moment where the idea had to prove itself.
So it leaves the room clean… but untested and weak.
And that’s where execution breaks.
The best ideas don’t come from speed.
They come from constraint, friction, and the willingness to stay in the part that isn’t clear yet.
AI is powerful because it creates order.
But the goal isn’t just to get to an answer.
It’s to create something people believe in enough to act on, make decisions around, and communicate with both clarity and emotion. It’s to engineer with strategy, based on ROI. And that still requires doing the hard thinking—together—before anything gets finalized.