AI Operator · June 27, 2026 · 3 min read
The Farmer Isn't Ashamed of His Tractor
The tractor never made a farmer anything other than a farmer, he produces better, that's all. The vertigo we feel right now doesn't come from the tool. It comes from a bad question. As long as you're asking "will I be replaced?", you've already lost.
There’s something telling about the way intellectual professions respond to AI. Surgeons don’t tremble before the electric scalpel. Bricklayers don’t write op-eds about the “existential threat” of the cement mixer. But the consultant, the copywriter, the lawyer, they question their survival. Not their effectiveness. Their survival.
The question is wrong. And a wrong question always produces a wrong answer.
The machine displaces, it doesn’t erase
The tractor doesn’t plant by hand for you. It displaces the farmer’s work: less physical effort on tillage, more decision-making on crop rotation, technical sequencing, reading the soil. The craft doesn’t disappear, it reorganizes around what the machine cannot do. The farmer stays a farmer. He produces better, that’s all.
AI works the same way. It displaces work upward: less mechanical execution, more framing, judgment, quality control. An LLM that drafts a first pass doesn’t eliminate the writer, it eliminates the part of the work the writer never liked in the first place. What remains is precisely what defined the craft before the pressure of output buried it.
The vertigo doesn’t come from the tool. It comes from a badly framed question.
Solutionism as trap
Evgeny Morozov named the problem with precision. In To Save Everything, Click Here, he describes technological solutionism: the tendency to reframe every complex human situation as a “problem” to which technology can provide a “solution.” The shift is subtle but fatal, because it presupposes the situation was a problem before anyone decided it was.
Applied to AI at work, the mechanism is identical. Start with “humans are slow, expensive, fallible”, and suddenly AI appears as the obvious fix. But that framing isn’t neutral. It has already decided what the problem is. It has already chosen what counts.
The real question isn’t “can the machine do what I do?” It’s: “what is my craft actually supposed to produce, and does the machine help me produce it better?”
Those are two different questions. The first leads to anxiety. The second leads to work.
Out of the defensive stance
“Will I be replaced?” is a survival question. It triggers a defensive, identity-driven, unproductive posture. It pushes you to protect the perimeter instead of redrawing it.
“What does my craft become?” is a practice question. It forces you to look at what you actually do, what holds value, what no longer does. It demands clarity.
Any operator building with LLMs learns this distinction fast. AI executes well what is well framed. It hallucinates where thinking is vague. It reveals, in negative space, the quality of the human framing upstream. That’s not a threat, it’s a diagnostic. The crafts that survive AI won’t be the ones that resist it. They’ll be the ones that know what they produce and why.
Redefine, don’t resist
Resisting the tool is a posture that burns energy and builds nothing. Redefining the craft around what the tool cannot do, that’s where the real decision lives.
The farmer who refused the tractor didn’t save his craft. He exhausted it. The one who adopted it could finally think the farm rather than merely endure it.
AI is no different. It frees up cognitive capacity, provided you know what to do with it. And that is precisely where the real professional question plays out: not “will I be replaced by the tool,” but “am I capable of defining what the tool doesn’t replace.”
The shame isn’t in using the tool. It’s in no longer knowing what you’re doing with it.
The letter
Analysis, reversals, tenable hypotheses.
SUBSTACK · NAKKAMA.SUBSTACK.COM