Every few years, a new wave of panic sweeps through the economy: "The robots are coming for our jobs!" This time, it's AI. Last time, it was automation. Before that, it was outsourcing. Before that, computers.
But here's the uncomfortable truth: AI didn't take your job. Your job was already obsolete.
The Myth of Job Stability
We've been sold a lie: that jobs are stable, permanent fixtures of the economy. That if we go to school, get a degree, and work hard, we'll have secure employment for decades.
This was never true. It was a brief anomaly in human history—roughly 1950 to 1990 in the West—when union power, post-war economic expansion, and manufacturing dominance created an illusion of job security.
The moment globalization and digital technology arrived, that era ended. Not because of a conspiracy, but because of simple economics: if a task can be done cheaper elsewhere (or by a machine), it will be.
What Really Happened
Your job didn't disappear because of AI. It disappeared because:
- It was already repetitive. If 80% of your work can be described in a flowchart, it's automatable. AI just made the automation cheaper.
- It was already commoditized. If your skills are widely available, employers will pay less for them or find alternatives.
- It was never designed for humans. Most office jobs were created to manage information before computers existed. We're still doing 1960s workflows.
The Real Problem
The problem isn't AI. The problem is that we've tied survival—food, housing, healthcare—to employment. And employment is increasingly optional for corporations.
This is why conversations about automation always feel dystopian. We're not afraid of machines doing work. We're afraid of losing our ability to survive.
What to Do About It
First, stop blaming technology. AI, automation, and software are tools. They're not good or evil—they're inevitable.
Second, recognize that the employment contract is breaking down. Waiting for it to "come back" is a losing strategy.
Third, start building alternatives:
- Automate your own work. If a task can be automated, automate it and keep the time savings.
- Build skills that can't be commoditized. Creativity, judgment, relationship-building, taste.
- Create alternative income streams. Digital products, services, micro-businesses.
- Advocate for policy changes. Universal basic income, worker ownership, right-to-disconnect laws.
The Opportunity
Here's the optimistic take: if AI can do the boring parts of your job, that's not a threat—it's liberation.
The question isn't "Will AI take my job?" The question is "What will I do with my time when the boring parts are automated?"
That's the future we should be building. Not one where we fight to keep pointless jobs, but one where automation frees us to do work that actually matters.
August 2026 update: the evidence showed up, and it was more specific than this essay
Editor's note, August 2026. This essay was published in March 2025, before there was much hard data on any of it. There is data now, and it sharpens the argument in one place and undercuts it in another. Original text unchanged above.
The claim here was that AI didn't take your job because your job was already obsolete. The first serious payroll evidence suggests something narrower and, honestly, worse: the damage is showing up mostly in hiring, not in firing.
Economists at the Stanford Digital Economy Lab — Erik Brynjolfsson, Bharat Chandar and Ruyu Chen — have been tracking millions of workers through ADP payroll records in a paper called "Canaries in the Coal Mine?". Their finding, updated through 2026: workers aged 22 to 25 in the most AI-exposed occupations have seen a double-digit relative decline in employment, while workers in their thirties and forties doing the same jobs have not. The mechanism isn't mass layoffs. It's firms quietly not opening the junior req.
So: half right. The essay was right that the exposed work was the routine, describable-in-a-flowchart kind. It was wrong to frame this as something that happens to your job. It's happening to the job you would have been hired into. The people who already have the experience are, for now, fine. The people who needed an entry-level rung to acquire that experience are the ones paying.
That changes the advice slightly. "Automate your own work and keep the time" still holds if you already have a job. If you don't, the harder problem is getting a first one at all — and the honest answer is that nobody, including us, has a good playbook for that yet.