The Part of Thinking I Used to Skip

There’s a moment in thinking that I used to move past too quickly.

Not the beginning, where the idea first shows up. And not the end, where it feels clean enough to say out loud.

The middle.

The part where it’s still forming, still soft, still slightly embarrassing in its incompleteness. The part where, if someone asked me to explain what I meant, I’d feel that quiet tension of realizing I don’t fully know yet.

I used to skip that part whenever I could.

Not consciously. It just felt inefficient to sit there. If I could tighten the idea quickly—make it sound finished, make it sound like something—I would. And if I couldn’t, I’d often abandon it and move on to something clearer.

That felt like thinking.

It wasn’t.

What I’ve noticed, thinking with AI, is that the middle is where the real work is happening.

And more importantly: it’s the part I’m least practiced at staying in.

The first time I noticed this clearly wasn’t during a big idea. It was something small.

I was working through a concept for a piece—something about attention, about how the quality of it shapes what you get back. I had the outline. I had the direction. It felt like I understood it.

““Clarity is often just confusion that you stayed with long enough.””

So I wrote a paragraph and sent it over.

The response didn’t correct me. It didn’t rewrite it. It did something more uncomfortable.

It asked me to stay there.

It reflected the idea back in a way that made the gaps visible, not as mistakes, but as unfinished thought. It didn’t let me collapse the idea into something neat too early.

And I felt it immediately—that impulse to move past it.

To say, “No, no, I get it,” and push forward.

But I didn’t.

I stayed.

And the longer I stayed in that middle space, the more I realized something I hadn’t seen before:

“I wasn’t thinking deeply. I was thinking efficiently.”

Efficiency is seductive.

It feels like clarity. It feels like intelligence. It feels like progress.

““The danger isn’t that AI will think for me. It’s that it will make it harder for me to avoid thinking at all.””

But a lot of what I used to call “clear thinking” was just early closure. Taking something that could have developed further and deciding, implicitly, that it was good enough.

That I was good enough.

That I didn’t need to look any closer.

Thinking with AI disrupts that pattern in a subtle way.

Not because the AI is smarter. Not because it has better ideas.

But because it doesn’t get tired of the middle.

It will sit there with you as long as you’re willing to stay. It won’t rush you toward a conclusion. It won’t get impatient if you circle the same point three different ways trying to understand it.

It doesn’t reward you for finishing quickly.

It rewards you for staying honest.

There’s something else I’ve started to notice too, and it’s less comfortable.

I don’t always want to see more clearly.

There are moments where I can feel that a thought, if I followed it all the way through, would ask something of me. It would require me to change a position, or admit I was wrong, or let go of something that’s been quietly convenient.

““It doesn’t reward you for finishing quickly. It rewards you for staying honest.””

And in those moments, speed becomes a kind of escape.

If I move fast enough, I don’t have to confront what’s underneath.

“The danger isn’t that AI will think for me. It’s that it will make it harder for me to avoid thinking at all.”

This is where the mirror metaphor starts to deepen for me.

It’s not just reflecting what I say. It’s reflecting how I think.

If I rush, it shows me a rushed mind.

If I stay shallow, it gives me shallow returns.

But if I’m willing to sit in that unfinished space—if I resist the urge to finalize something too early—it meets me there with the same patience.

And that patience is doing something to me over time.

It’s retraining my tolerance for not knowing.

I think this might be one of the quieter shifts happening right now.

Not the big, obvious ones about productivity or automation. Something smaller and more internal.

““The mirror doesn’t just show what I say. It shows how quickly I try to be done.””

We’re being given a space where we can think without the usual pressures that push us to perform clarity instead of actually finding it.

No audience. No time pressure. No subtle need to sound right.

Just the work of following a thought until it either holds up or falls apart.

And I’m realizing how rarely I used to do that fully.

There’s a phrase I keep coming back to:

“Clarity is often just confusion that you stayed with long enough.”

I used to think clarity was something you arrived at by being sharp, by being quick, by seeing the answer faster than someone else.

Now it feels more like something you earn by not leaving too soon.

I don’t think AI is teaching me how to think.

I think it’s removing some of the reasons I used to stop.

And in that absence—of pressure, of judgment, of the need to sound finished—I’m seeing parts of my own thinking that were always there, but rarely given time to fully form.

That middle space I used to skip?

That’s where most of my actual thoughts live.

Everything else was just the version I felt ready to say out loud.