Who Are You When the Work Takes Seconds?
The machine took the task. It cannot take the caring, and the caring was you the whole time.
There is a particular feeling that arrives the first time you watch a machine do, in four seconds, the thing you spent years becoming good at. It is not quite anger, and it is not quite fear. It is closer to vertigo. A small, disorienting drop in the stomach, and underneath it a question you did not expect to have to answer. If the thing took seconds, then what, exactly, was I?
We built ourselves into the doing
For most of us, what we do quietly became who we are. We say "I am a writer," "I am an engineer," "I am the person who can do this difficult thing," and we do not mean only that it is our job. We mean it is our self. The skill was not just labor we performed; it was an identity we earned, slowly and at cost, and then wore like a name. You could not have the output without first becoming the person.
The machine pulls them apart
And now a machine produces the output without ever having become the person. It does not just do the task. It quietly unbundles the task from the identity that used to be inseparable from it. The thing you thought required someone like you turns out to require, apparently, no one at all.
The machine can produce the output without ever having been the person. That severance is the vertigo.
The equation was always a little false
But it is worth asking whether the welding was ever quite right. We made the having-done the source of our worth. We let productivity stand in for identity, and so when productivity becomes cheap, identity feels cheap along with it. The logic is tidy and, I think, mistaken. The output was never actually the self. The output was the evidence we pointed to when someone asked who we were, because it was easier to point at a thing than to explain a caring.
It produces results without wanting them to be good. The wanting, the minding, the stubborn insistence that this particular thing deserves to be done well, that was always the actual you.
You were never your deliverables. You were the one who gave a damn about them.
I am what I produce. Output equals worth. So when the output gets cheap, I must be getting cheap along with it.
The output was only ever the evidence. You were the caring underneath it, the wanting it to be good. That, the machine cannot make.
"This sounds like a consolation prize"
The sharp objection: telling someone their output is not their worth is exactly what you say to a person who just lost the ability to produce anything valuable. It can be a cope, a soft blanket thrown over obsolescence. People take real, legitimate pride in what they make. Why should they surrender that just because a machine can now make a passable version too?
They should not surrender the pride. The point is narrower and older than AI: welding your whole identity to your output was always a fragile bet, and we knew it before any model existed. Look at elite athletes at retirement. Sports psychologists have a name for what so many of them fall into, athletic identity foreclosure, where a person built so completely around performing collapses the moment the performing stops, even though nothing about their actual worth changed. The body was always going to age out. The machine just hands the rest of us that same reckoning early. This is not consolation for the defeated. It is a correction that frees you: keep the pride in the craft, but stop letting the deliverable be the whole of you, because the deliverable was never reliable ground to stand on.
The question that was always yours
So the cheapening of the work forces a harder and, in the end, freeing question. If what you do can no longer tell you who you are, you are thrown back on the older question we have been avoiding for a very long time. Who are you when you are not producing anything at all?
It is a frightening question. It is also the only one that was ever really yours, and the machine, for all it can make, will never be able to answer it for you, or take it away.
I write about AI, data, and learning regularly at pinaldave.com, and I have been teaching this hands-on in my AI workshops.