Every office has a holdout on AI right now, and the holdout is often the best person in it. The bookkeeper who won’t touch it. The veteran who tried it once, watched it be confidently wrong about something they knew cold, and closed the tab feeling pleasantly vindicated. Maybe the holdout is you. The arguments against using this stuff arrive with real heat behind them, and before I argue back I want to name them properly, because they are better arguments than the people selling AI tend to admit.
The first one is pride, and it’s legitimate. You got good the long way: years of doing the work slowly, then badly, then less badly, until the judgment finally moved into your hands and stayed. Now a box on the internet turns out a draft of that work in under a minute, and everyone around you grins like this is simply wonderful news. Nobody grieves that out loud in a Tuesday meeting, but it’s grief, and pretending it isn’t is how you lose the room.
The second one is self-expression, and it’s legitimate too. The proposal reads like you because you wrote it, the client can tell, and some mornings that’s the part of the job that feels least like work. Handing it to a machine feels like signing your name under somebody else’s handwriting.
Both objections deserve a straight answer instead of a pep talk. Mine carries a little edge, so let me say where the edge points before I swing: at the framing, never at you.
The framing says using a tool is somehow wrong. Say that sentence out loud and listen to it, because I have turned it over a lot and I cannot find the wrongness anywhere in it. We ran this argument once already, at full national scale, over the horse. Being good with horses was real skill and real identity. People were proud of their teams, and the pride was earned, and the car arrived carrying no opinion about any of it. The car was just faster. That is the entire thing the car was. Faster.
Nobody takes pride in walking to work. A few people walk for the exercise, and respect. But nobody walks for the pride of it in pouring rain, and nobody at all at zero degrees with the wind coming sideways, because out on that sidewalk there is no craft being honored and no self being expressed. There is just a cold, wet person who could have driven. Speed is not a character flaw. It never was.
So retire the enemy framing; it’s costing you information. Treat this thing the way you have treated every system that ever showed up at your door: the new billing software, the phone tree, the first truck that came with a computer in it. Sit down with it on a slow afternoon. Feed it a problem you already know the answer to, and watch closely, because it will be strong in places that surprise you and it will make things up with a straight face, sometimes in the same paragraph. Learn its tells. Decide what it is allowed to touch in your shop and what it is never allowed to touch. That is what an operator does with a system. Figure out what makes it tick, then put it to work on your terms.
Notice what stayed yours the whole way through: the judgment. The tool drafts, you decide. It suggests, you sign. The pride you spent twenty years building moves up a floor, from producing the draft to being the one person in the room who can tell a good draft from a convincing bad one, and that skill just got more valuable. Your voice survives too, because the last pass is where the voice lives, and nothing about the tool takes the pen out of your hand on the last pass. You would have to set it down yourself.
This shop got started on exactly that itch, taking systems apart to see what made them tick, back when the systems had fewer manners. Same curiosity, older machines, and a story for another page. For now the invitation is simple. Bring the pride; it is the qualification. Bring the suspicion too, honestly, it stays useful. Point the curiosity at the tool instead of at the argument, and see what you find. And if your office is mid-argument about all of this and could use a third chair for an hour, talk to us.