The Great Tech Anxiety
American workers are panicking. Polls show long-term employment optimism has plummeted to historic lows. The average worker now believes they face a 22% chance of losing their job within five years—a level of dread higher than during the 2008 financial crisis.
Even Silicon Valley’s elite are fueling the fire:
Dario Amodei (Anthropic): Warns AI could drive unemployment to 10–20%.
Bill Gates (Microsoft): Claims humans won't be needed for "most things."
Sam Altman (OpenAI): Warns of "significant disruption" as the world transitions to new roles.
Yet, despite the hand-wringing, the actual data tells a completely different story. The OECD employment rate is breaking records, rich-country unemployment sits at a tiny 5%, and America is employing more people than ever in "AI-exposed" fields like law.
Why Economists Aren't Panicking
Mainstream economists reject the "lump of labour fallacy"—the mistaken idea that there is a fixed amount of work to go around. When technology automates a task, it lowers costs and enriches consumers. That saved cash is then spent elsewhere, creating entirely new industries and jobs.
If AI genuinely triggers mass, long-term unemployment, it will be the first time in human history that technology has ever done so. History shows that job disruption is kept in check by two massive structural brakes:
1. Tech Diffusion is Painfully Slow
We overestimate how fast the world changes. Historical data reveal that frontier economic growth has never sustained a pace above 2.5% a year. Innovation takes generations to scale.
The Tractor: Invented at the dawn of the 20th century, it took decades—not years—to significantly shrink the agricultural workforce.
The Computing Boom: In the mid-20th century, computers and shipping containers disrupted the workforce at twice the rate we see today. The result? A golden era of rising wages and booming opportunities.
2. The "Industrial Revolution" Myth
AI doomers constantly point to the 19th-century Industrial Revolution—and specifically "Engels’ Pause"—as proof that machines starve workers while enriching capitalists.
Modern economic scholarship has thoroughly debunked this narrative:
Job Creation: Between 1760 and 1860, the British workforce more than doubled (from 4.5m to 12m) while unemployment remained low.
The Real Villain: Wages stagnated not because machines took the jobs, but because productivity growth was slow and politicians spiked the cost of living with food tariffs and wartime inflation. Employers paid fairly; government policy failed.
"The Industrial Revolution is not a template for technological change that boosts productivity at the expense of labour."
— Nicholas Crafts, Economic Historian
What to Watch For Next
Could this time actually be different? Perhaps. But if a true AI jobs apocalypse is coming, it won't hide in opinion polls. It will flash red in the economic data through three specific signals:
Surging Productivity: US GDP per person breaking past the historical 2.5% ceiling.
Profit Decoupling: Corporate profits are skyrocketing while real wages stagnate.
The Recession Cleansing: Mass, permanent layoffs in white-collar sectors during the next economic downturn.
Until those metrics move, the tech titans and pollsters are just guessing. History bets on the workers; the machines will have to wait.