On AI, the illusion of prompt engineering, and the difference between writing faster and writing truer.
Right now, your feed is flooded with guides on “how to use AI without sounding like AI.”
I use AI practically every day for research, structuring, and writing. I read the guides. Most of them are just tactical tricks to avoid sounding robotic. But fixing the sentence structure doesn’t fix the core problem. The output is still predictable.
What most people miss when using AI is this: You must own the thinking before you touch the prompt bar.
Here are the five practices to ensure the machine works for you, not the other way around.
1. OWN THE THINKING Arrive with your angle before AI touches anything.
AI executes. It doesn’t lead. Before you open any AI tool, you should already know: What is the point I am trying to make? What do I actually believe about this? What would most people assume—and where do I disagree?
If you can’t answer those questions before you start, AI will answer them for you. And its answer will sound exactly like everyone else’s.
The test: If you removed the AI from what you produced, is the thinking still yours?
2. READ THE ROOM Check the moment before you check the message.
The right idea at the wrong moment lands flat. The same idea, timed to a conversation already happening, lands like a confirmation of something people were already feeling.
Before writing, ask: Who is this actually for? What are they feeling right now? Is this conversation already in the room, or am I starting it from scratch?
This is not about chasing trends. It is about genuine situational awareness—the difference between speaking into silence and speaking into something that is already alive.
The test: Would this land the same way if posted tomorrow? If yes, it’s not reading the room. It’s just posting.
3. FIND THE PATTERN (AND THE GAP) Spot what others have accepted as normal.
The most useful observations are rarely new information. They are existing information, seen from a different angle.
Most LinkedIn profiles look the same. Most AI outputs use the same five sentence structures. Most corporate content says the same things in the same order. The pattern is already there; most people just stopped noticing it. Your job is to name what others have made invisible by accepting it as the default.
Then, look for the gap: What angle is everyone avoiding? What is the behavioral friction behind it? Is there a tension that existed long before the current crisis? If everyone is shouting on one side of the aisle, the real insight is usually sitting quietly in the gap.
The test: Could this observation have been written by anyone, or does it require someone who was actually paying attention?
4. STAY GROUNDED Anchor everything to lived experience. Never invent.
AI will invent confidently. It will produce statistics that don’t exist, quotes that were never said, and anecdotes that never happened—all in fluent, authoritative prose.
Your protection is simple: If you cannot trace a claim back to something real—something you saw, did, verified, or experienced—it does not go in.
This is not just about facts. It is about voice. The moment you allow AI to invent an experience you haven’t had, the writing stops sounding like you. Readers feel it before they can name it.
The test: Can I point to the real moment, source, or experience behind every claim in this piece?
5. SLEEP ON IT Let the work breathe. Cut what doesn’t hold on second look.
AI produces fast. That speed is useful and dangerous at the same time.
The discipline is to not use the output immediately. Leave it. Come back the next day, or even a few hours later, with fresh eyes. Ask: Does this still hold? Is this actually true, or does it just sound right? What would I cut if I were reading this for the first time?
What survives that second look is what belongs in the work. What doesn’t survive it was never yours to begin with.
The test: Does this read the same way today as it did when I wrote it yesterday?
The Sequence
These five practices follow a specific order—and the order matters.
Inside → Out → Inside
You start from within (Own the Thinking), move outward to read the context (Read the Room, Find the Pattern, Stay Grounded), and then return inward to edit with judgment (Sleep on It).
This movement is not a checklist. It is a habit of mind—one that, practiced consistently, becomes naluri (instinct).
What This Is Not
This is not a prompt library or a productivity system. It is not about writing faster.
It is about writing truer — and knowing the difference between the two.