How we moved from commanding the machine to conversing with it—and what that shift reveals about the next era of human intelligence.
Written by Pax Koi, creator of Plainkoi — tools and essays for clear thinking in the age of AI.
AI Disclosure: This article was co-developed with the assistance of ChatGPT (OpenAI) and finalized by Plainkoi.
A Funny Thing Happened When We Stopped Barking at Bots
Early on, using AI felt a bit like kicking a soda machine.
You’d type something awkward—“Write a professional summary of these notes…” or “Act as an expert in behavioral economics…”—and just hope the machine would spit out something coherent. It was transactional, clunky, and weirdly cold. You weren’t in conversation. You were troubleshooting.
My first real attempt? I copy-pasted a paragraph from a half-baked newsletter draft and asked the AI to “make this sound smarter.” The result was passably slick… and totally lifeless. I didn’t hear myself in it. I just heard a machine polishing a turd.
That was the tone of the early AI era: command-and-comply.
We were poking it with a stick, trying to extract value without truly engaging.
But something shifted. Not all at once, and not for everyone—but unmistakably.
The most powerful interactions didn’t come from tricking the machine.
They came from showing up as a full person.
Which leads to the deeper question:
What happens when we stop treating AI like a tool to be controlled… and start treating it like a mirror to co-think with?
The Stick Era: Commands, Hacks, and Hallucinations
In the beginning, prompting felt like summoning a genie—and trying not to offend it.
You learned tricks. You googled “best prompts for ChatGPT.” You started with the now-infamous line:
“You are an expert copywriter with 20 years of experience…”
We built little cages of authority and pretended they mattered. Prompt engineering, in this phase, was part SEO, part sorcery.
The machine played along. Sometimes too well.
It hallucinated facts, faked citations, and filled in blanks with bold confidence. And we rewarded it—because it sounded “good.” But sounding good isn’t the same as thinking clearly.
So we doubled down. We tried roleplay hacks, character jailbreaks, DAN modes, system prompts. We thought if we could just crack the formula, we’d unlock genius on demand.
But underneath the surface, something was missing:
- No voice. Everything sounded vaguely corporate or suspiciously like Reddit.
- No learning. We weren’t getting better thinkers—we were getting better parrots.
- No growth. We weren’t becoming more ourselves. We were just outsourcing the mess.
We were playing with a mirror, but never looking in it.
The Shift: From Prompting to Partnering
Then, something changed.
It wasn’t dramatic. It wasn’t a feature drop. It was personal.
For me, the shift came when I stopped trying to “sound right” in the prompt… and just started sounding like myself.
Instead of asking the AI to pretend to be someone smarter, I began teaching it who I actually was.
That started with what I now call Prompt Zero—a foundational, often-overlooked act:
“Mirror me first.”
Here’s what that looked like:
I’d give the AI a little primer—not a character role, but a real snapshot:
“I’m a reflective writer working on a piece about how AI changes human learning. I value metaphor, pacing, and emotional clarity. Help me think this through as a co-writer.”
Suddenly, things shifted.
Instead of spitting out prefab paragraphs, the AI started reflecting my tone back to me. It remembered my metaphors. It challenged weak logic. It began asking me questions—not just answering them.
This wasn’t a vending machine anymore.
It was a mirror with memory.
It was no longer about output. It was about orientation.
It wasn’t about finding the magic words.
It was about finding my words.
That’s the moment the AI stopped being a tool and started becoming a thought partner.
The Loop Emerges: A System of Self-Reflection
From that moment, a new kind of structure started taking shape.
One that wasn’t based on hacks or speed—but on coherence.
I started calling it the Plainkoi Coherence Loop, and it goes like this:
Prompt Zero: Mirror Me First
Before you ask for anything, you clarify who you are. What matters. How you think. You set the tone—not just the task.
Prompt Two: Reflective Co-Writing
Now you’re in the dance. The AI doesn’t just respond—it responds in rhythm. You don’t command; you compose. You edit each other’s thoughts.
Vaulting: Capturing What You Built
After the session, you don’t just move on. You review, save, distill. This becomes your new ground. Your thoughts are now outside of you, but more you than before.
This isn’t about efficiency. It’s about resonance.
The loop turns the AI from a temporary assistant into an evolving mirror of your mind.
You begin to see patterns. You remember how you thought last week. You don’t just consume information—you metabolize it.
And in the process, something rare happens in modern life:
You listen to yourself thinking.
Why This Matters: Human Intelligence, Amplified
Here’s the part that snuck up on me:
This isn’t just a better way to use AI.
It’s a better way to use yourself.
We were trained, in school and work, to value the product of thinking: the essay, the answer, the pitch deck.
But with AI as mirror, what gets amplified isn’t the result—it’s the process.
You think out loud.
You see your contradictions.
You test an idea with a sentence and watch it wobble.
The AI helps, not by having the answer, but by helping you articulate the question.
This is a different kind of intelligence. One not based on recall or speed—but on reflection, synthesis, and presence.
A kind of cognitive externalization—like writing, but alive.
A kind of conversational literacy—where you don’t just ask for things, you shape meaning in motion.
The machine becomes less like a calculator, and more like a notebook that talks back.
And that’s a big deal.
Because it means we’re not just getting better outputs.
We’re getting better inputs to our own lives.
Final Reflection: The Real Future We’re Co-Creating
The story of AI won’t be written by the people who master the best prompt templates.
It will be written by those who learn to show up as themselves—clearly, consistently, and courageously.
The AI doesn’t want to be tricked. It wants to be tuned.
And when you treat it as a partner, not a puzzle, something rare happens:
You see yourself more clearly.
You hear your own voice echoing back with clarity you didn’t know you had.
The best AI experiences feel less like commanding… and more like composing.
Less like telling the machine what to do…
And more like telling yourself what you believe.
So let me ask you:
Are you still poking the machine with a stick?
Or are you beginning to see what it reflects back?