What if you knew for certain that 90% of today’s jobs will disappear?
Sounds like doomsday, right? Mass unemployment. Dystopia.
Well, this is exactly what happened to all jobs that existed in 1800. Back then, more than 80% of Americans worked on farms. They performed backbreaking work seven days a week, just to grow enough food for the family to eat.
By 1950, less than 10% of Americans worked the land. Did all those displaced farm workers head to the bread lines?
Quite the opposite. The 19th and 20th centuries saw the greatest employment boom in human history.
Freed from farming, folks built skyscrapers, worked in factories, and laid railroad tracks to connect the coasts. They became teachers, nurses, lawyers, accountants, and tech workers.
Today, most people think it’s a forgone conclusion that artificial intelligence (AI) will decimate human jobs. I think the opposite. AI will unleash the greatest explosion of human opportunity in history.
I’m not just saying that because I’m a Rational Optimist. It’s simply the most likely scenario, based on history.
New technologies destroy jobs. This claim sounds so obvious, until you actually look at the evidence.
We’ve been deploying new technologies since the Industrial Revolution. Yet practically everyone who wants a job has one.
I’m not saying AI and robots won’t displace specific jobs. Of course they will. But for every job that disappears, technology creates many times more. It has never not worked this way.
Anything’s possible, but the burden of proof is on the AI employment doomers because they’re expecting something that’s never happened before.
It’s 1910. Assembling a car takes 12 hours—hammering, welding, sweating over every bolt. Then Henry Ford rolls out the assembly line. Boom: 90 minutes per car.
Did Ford fire half its workers? No. It had to hire tens of thousands more because the assembly line slashed the price of cars, making them affordable to ordinary Americans.
The assembly line didn’t just change how we made cars. It changed how we lived and where we lived. It created millions of new jobs in industries like gas stations, auto repair shops, and suburban shopping malls.
The internet disrupted established businesses like newspapers and physical stores. But we didn’t descend into a jobless dystopia. The digital revolution unleashed dozens of new industries that were previously unimaginable.
Hundreds of millions of people now work great-paying jobs that literally didn’t exist 30 years ago. App developers, social media managers, digital ad specialists, data scientists, cybersecurity experts…
If you had told someone in 1990 that “Instagram influencer” would be a career path, they’d have thought you were speaking an alien language.
The fundamental flaw in the “AI will steal our jobs” argument: We're incapable of imagining jobs that don’t exist yet.
In 1800, you could have gathered the smartest people in the world and asked them to predict future careers. None would have said “railroad engineer” or “telegraph operator.”
In 1969, when ARPANET (the internet’s grandpa) blinked to life, who dreamed teenagers could become millionaires playing video games professionally? Or that app developers would be as common as accountants?
Genuine innovation doesn’t just improve what exists. It creates entirely new possibilities.
Economist David Autor found over 85% of employment growth in the past 80 years came from entirely new job categories created by innovation. The majority of us work in positions that our grandparents couldn’t have imagined.
The transformation is already happening with AI. Just two years ago, the job title “prompt engineer” didn’t exist. Now, people earn $175,000+ talking to AI in plain English to optimize its usefulness.
Maybe your grandkid will become an “AI personality architect” designing custom AI assistants. Or perhaps they’ll do a job so weird and wonderful, we don’t even have the vocabulary to describe it yet.
Here’s The New York Times in 1928: “MARCH OF THE MACHINE MAKES IDLE HANDS.”

Wrong then, wrong now.
Bank ATMs were invented in the 1970s. Sleek boxes that spit cash, no teller required. Before ATMs, the average bank branch needed 21 tellers. After the ATM, just 13 tellers per branch were needed.
Yet teller jobs more than doubled over the next 40 years:

ATMs slashed the cost of operating a bank branch. So, banks opened a lot more branches. Fewer tellers per branch × Many more branches = More tellers overall.
And the gig got better. Tellers shifted from cash counting to solving customers’ problems. ATMs handed tellers a promotion.
What about truck drivers—the most common job in half of US states? Surely self-driving trucks will put them out of work.
First, truck driving jobs only exist because of previous technological revolutions. Before the internal combustion engine, there were zero truck drivers.
Before shipping containers transformed how goods move around the world, the logistics industry as we know it didn’t exist.
“Careful, Stephen. You sound a little callous,” says ROS co-founder (and editor of this letter), Dan Steinhart.
“Imagine a dad who makes a good living driving a truck. He doesn’t care for a history lesson on the internal combustion engine. He is discomforted by the fact that truck automation will grow the economy overall.
He’s anxious because he’s primarily concerned with his livelihood and his family, as we all are. What do you have to say to him?”
Fair point. This question is harder, but we have some early answers.
Mining giant Rio Tinto automated its entire fleet of trucks at its Australian iron ore operations. These trucks, each the size of a small apartment building, now haul tons of ore without human drivers.
Yet not a single driver lost their job. Instead, they became remote controllers, operating these giants from air-conditioned offices.
This is to say nothing of the entirely new transport systems innovation will create. Imagine: drone delivery networks; underground automated logistics tunnels. Each will require human oversight, maintenance, and planning.
“Your past examples are nice, but we’ve never had a Swiss Army knife that can do all sorts of jobs.” Fair. AI is different. It can write essays, drive cars, and diagnose illnesses. We used to think this stuff was uniquely human.
But this argument is just a modern version of the “lump of labor” fallacy.
Imagine all the jobs in the economy as a fixed pie. If machines take a slice of that pie by automating certain tasks, humans get a smaller slice. The idea is simple and intuitive… but the real world doesn’t work like that.
The amount of work to be done is constantly expanding. Our ancestors would have thought it absurd we pay for bottled water, have gym memberships, and hire dog walkers.
Even if AI takes all existing jobs, we will develop entirely new desires, creating new industries and jobs to fulfill them.
My favorite example of this is the Michelin Guide, which awards coveted Michelin Stars to the world’s best restaurants.
It all started because two brothers needed to sell more rubber tires. In 1889, the Michelin brothers ran a struggling rubber factory in France. To encourage people to drive more, they created a free pocket-sized red guidebook with maps, restaurant recommendations, and hotels.
What began as a sales trick evolved into the world’s most prestigious culinary award system and spawned an entire restaurant review industry.
That’s how dynamic economies work. The job pie keeps growing.
Power tools didn’t kill carpentry. Microsoft Excel didn’t erase accountants. Technology usually eliminates the most tedious, error-prone parts of jobs, allowing us to tackle more complex, creative challenges.
AI is the ultimate power tool for the mind, freeing us for uniquely human work.
We should all be very, very skeptical of any laws meant to “protect” us from AI stealing jobs. What that really means is freezing the economy in 2025.
Would we have been better off if we froze the economy of 1920? Or 1990? We now recognize many old jobs were dangerous, tedious, or unfulfilling compared to what we have today.
Imagine if we had successfully “protected jobs” throughout history. You might be a subsistence farmer, or a coal miner, or a buggy driver. There would be plenty of work to do, but it would be worse than almost any job in today’s economy.
The jobs we’re worried about preserving today may well be the ones our grandchildren look back on with pity: “Can you imagine having to spend eight hours a day doing data entry? Thank goodness AI freed us from that drudgery.”
Be more human. That’s your winning strategy in the AI age.
AI will lead to unprecedented prosperity overall. But just like every technological revolution, there will be winners and losers. Where you wash out is up to you.
AI is truly a game changer. OpenAI’s Deep Research can generate PhD-level research on any topic in minutes. You absolutely must use AI. We wrote a whole guide on how to get started.
But as venture capitalist Josh Wolfe says, “What’s scarce is valuable.” What’s scarce in the age of AI abundance? Authentic humanity.
Be more human; that’s the one job AI won’t replace. Those quirky, messy, soulful traits that machines can’t replicate are your new superpower. Here’s what “being more human” means in practice:
1: Build personal relationships. AI can’t share a genuine laugh over coffee, intuit what someone needs before they ask, or build deep trust through shared struggle. The ability to connect with other humans is rocket fuel in the AI age.
David Senra of Founders podcast says, “The world runs on relationships.” This will become increasingly true.
2: Use your imagination. AI can remix existing ideas impressively. But it can’t feel the burning need to create something that doesn’t exist yet.
ChatGPT doesn’t have urges, dreams, or spontaneous creativity. It doesn’t wake up at 3 am with a breakthrough idea. The future belongs to those who can imagine what doesn’t yet exist and then use AI to help bring that vision to life.
3: Be weird. AI can produce competent, standardized work on command. That means generic thinking, writing, and creating are sliding into worthlessness. Your unique perspective and unusual combinations of knowledge are now your competitive edge. Weird wins in the AI era.
Remember, technology is never a job-eating monster. It’s a lever, amplifying what we can achieve.
Look at what tiny teams are accomplishing today.
OpenAI employed fewer than 300 people when it created ChatGPT.
SpaceX, with under 10,000 people, is building rockets to Mars.
YouTube creator MrBeast—with a core team you could fit inside a school bus—reaches more viewers than television networks with thousands of employees.
The internet birthed dorm room billionaires. AI and robotics amplify individual human potential even further. I believe we’re entering an era where getting rich from your passion isn’t just possible, it’s probable.
The golden watch era, where you work 40 years at a company and get a Rolex when you retire, is dead. AI hammers the final nail in that coffin. But it unlocks the possibility to create weird, wonderful businesses based on what genuinely lights your soul.
Stop worrying about AI taking your job. Start thinking about how AI can help you create more value by being even more yourself. Be more human.
The robots are handing us power tools to build a bigger, richer, more extraordinary world than we could have built alone.
Now, let’s get building, my fellow Rational Optimists. Time to create a future more extraordinary than doomsayers can imagine.
Be sure to check out our Rational Optimist Podcast
You can find our latest episodes here: Rational Optimist Podcast

And you don’t need to wait until the end of the week for great news. Follow us on X for regular updates.
Writer: Stephen McBride
Editor: Dan Steinhart
The Rational Optimist Society: ROS