Let me put two numbers next to each other.
Around $120,000. That's roughly what a four-year degree costs.
Around $68,000. That's roughly the average salary for a new graduate.
Run the math and it takes something like five years just to break even, and that's before you count the interest on the student loans. Now layer on the part nobody wants to say out loud at graduation: a meaningful share of the jobs those degrees were supposed to lead to may not survive the decade. Forecasts vary a lot, but some put the number as high as 45% of jobs displaced by AI within a few years.
And it's not only the jobs people always assumed were at risk. It's not just factory lines and long-haul trucking. AI is coming for finance work. For marketing. For legal work. It's reaching into roles we were told were permanently safe, including doctors and engineers.
Sit with that for a second. You could spend four years and well over a hundred thousand dollars earning a credential, only to watch a machine do the job it qualified you for, faster and cheaper, before the loan is even paid off.
It sounds bleak. It isn't, actually. Because we've seen this movie before, and the ending depends entirely on which character you decide to be.
The Industrial Revolution, take two
Go back about 150 years. Roughly 95% of Americans were farmers. It was brutal, back-breaking work, and for most people it was nearly impossible to build any real wealth doing it.
Then the Industrial Revolution arrived and machines started replacing human labor. Most people had to pivot or get left behind. But a specific group saw what was coming and adapted. They didn't fight the machines. They built businesses on top of them. They treated the new technology as leverage instead of a threat, and they pulled away from everyone still trying to out-work a steam engine by hand.
We're standing at that same fork right now. The Industrial Revolution replaced manual labor. The AI revolution is replacing knowledge labor. The mechanism is different, the choice is identical: adapt and use the tool, or cling to a model that's quietly being dismantled underneath you.
Here's the part that should change how you read all of it.
AI is replacing employees, not owners
That distinction is the whole thing.
AI is astonishingly good at doing the work an employee does. It is not a substitute for the person who owns the asset that the work happens inside. It compresses the cost of building things. It does not replace the builder who captures the upside.
Which means the most important question in this whole conversation isn't "will AI take my job." It's "does my income depend entirely on a job in the first place." If every dollar you earn requires you to personally show up and trade an hour for it, you're in the category AI is built to disrupt. If some of your income comes from assets you own, you're in the category it can't touch the same way.
This is the same realization I had years ago about real estate, just pointed at a new asset class. A rental property is something you set up once, put systems around, and it pays you month after month whether or not you're at a desk. That's not a job. That's a cash-flowing asset. The rent doesn't care whether the labor market is healthy.
Software turns out to work exactly the same way, and AI just made it dramatically more accessible. We built and launched a mobile app in 30 days. Not years, not millions in venture funding, one month, with AI doing a huge amount of the heavy lifting. Then, mostly to prove the point to myself, I had an entire second software project built almost entirely by AI in two weeks. The same framework I use to evaluate a rental, build it once, systematize it, let it pay you, now applies to products that cost a fraction of a down payment to create. I wrote more about why a real estate investor would bother building apps in this piece, because the logic surprises people until they see it.
The barrier to becoming an owner has never in history been lower than it is right now.
What to actually do about it
I'm not telling you to drop out of school. I'm telling you that "get a degree and assume the job will always be there" is, in this environment, a genuinely risky single-point plan. The people who thrive through this shift will be the ones who learn to generate recurring revenue, through real estate, through online business, through automation. The people most exposed are the ones whose entire financial life hangs on a paycheck someone else controls.
So the move is to start building income that works for you instead of income you have to keep showing up to earn. Income that pays you every month. Income an algorithm can't simply absorb. You don't need a four-year degree or six figures of debt to begin. You need to make the shift from thinking like an employee to thinking like an owner, and then build the first asset.
If you want the mindset side of that shift, I dug into it in the inner game of wealth, and if you want the practical on-ramp to income that isn't tied to your location or your boss, start with how to build location-independent income.
Every major technological shift in history has been a fork in the road. One path adapts and builds wealth the way the industrialists did when the machines arrived. The other clings to a system right as it starts to give way. AI didn't create that fork. It just made it impossible to keep ignoring.
The reassuring part, and the part worth sitting with this week, is that the response isn't to outrun the technology. It's to own something the technology works for instead of against. Pick one asset, real estate or a small software product or any business that earns while you sleep, and start building it. That's the entire difference between the people this revolution crushes and the people it makes rich.
