Rebuilding Identity in the Age of AI
When the marginal cost of producing goods and services trends toward zero, capitalism’s pricing mechanism breaks down.
When knowledge work — software development, legal counsel, financial accounting — gets pushed from “scarce” to something as cheap as food, ordinary people go through the painful experience of losing their sense of self.
You are the “end,” not the “means.”
As AI becomes more and more embedded in everyday life, I’ve been coming across some interesting signals lately about how people are rebuilding their identities.
When the marginal cost of producing goods and services trends toward zero, capitalism’s pricing mechanism starts to fall apart. And as AI gradually pushes knowledge work — software development, legal advice, financial accounting — from “scarce” to something as cheap as food, capital will likely chase whatever remains scarce: physical presence, land, real experience. Faced with this shift, a lot of ordinary people are quietly going through the pain of watching their sense of self dissolve.
When a software engineer’s self-image — “I’m smart, I solve hard problems” — gets shattered by an AI that solves those same problems faster, what collapses isn’t just professional pride. It’s the whole structure of daily life. Because work isn’t just a source of income. It’s ritual, routine, the backbone of how we organize our days.
But things always cut both ways. Technological change also opens up the possibility of moving up. We have to actively detach from our old identities. Because the identities we’ve carried were built around tasks — “I’m a programmer who knows Python and C++.” When AI writes code a hundred times faster than you, that identity falls apart, and the panic that follows is real.
What you need instead is a new identity: “I’m a problem-solver who uses technology to make organizations run better.” From that position, AI writing code fast doesn’t destroy you — it becomes your leverage. Your value lies in spotting where things are inefficient and then directing the AI to fix it.
At that point, you are the “end” — what problem do I want to solve — not the “means” — what is it that I do. Taste becomes the core competency. Knowing what’s good matters more than being able to make it yourself.
Something else worth mentioning is leadership. Traditionally, leadership meant breaking down tasks and supervising execution. The boss sets the direction, middle management translates it into goals, and the people at the bottom grind it out. But as I said, AI has already brought the friction of “executing tasks” down to zero.
If you were to announce some wildly ambitious goal today — say, “we’re rebuilding the entire department’s workflow with large language models” — AI can produce a perfect plan, complete with a code architecture and execution path, in minutes. The real bottleneck isn’t the how. It’s the human friction of getting people on board.
What we actually need to do is guide the people around us, step by step, through every piece of outside information, every engineering insight, every milestone — slowly adjusting how they see the world. If you announce a massive change out of nowhere, people get confused and resistant. But if you’ve been laying the groundwork all along, when the change finally comes, it just feels inevitable. People don’t just go along with it — they look at you and ask, “What took you so long?”
This way of thinking has helped me get clearer on what actually matters in a world that keeps shifting. I hope it’s useful for you too.
Source: #494 – Jensen Huang: NVIDIA – The $4 Trillion Company & the AI Revolution The displacement of cognitive labor and what comes after