The New Luddites Are Coming — And This Time They Have a Point

Nottingham, England, 1811. Textile workers broke into factories at night and smashed the machines. History called them Luddites and filed their rebellion under “failed.” Case closed. Except now, in 2026, the tension feels eerily familiar. AI is coming for white-collar jobs so fast that “who’s going to smash the machines this time?” has stopped being a joke.

The Numbers Are Alarming

In 2025 alone, global tech companies announced AI-related layoffs exceeding 150,000 workers. Call centers, translation, junior coding, graphic design — job categories once considered safe fell onto the automation chopping block one after another. The IMF has warned that roughly 60% of jobs in advanced economies sit within AI’s blast radius, and that estimate keeps getting revised upward.

The real problem is speed. During the Industrial Revolution, mechanization took decades to displace craft labor. AI threatens new job categories every few months. When people aren’t given time to adapt, you get kindling. All it needs is a spark.

The Warning Signs Are Already Here

Nobody’s storming data centers yet. But the signals are hard to miss.

The 2023 Hollywood writers’ and actors’ strike was the first major collective action against AI replacement. Since then, labor disputes over AI have multiplied across industries and continents. In 2025, protests against autonomous taxis erupted in several European cities, with some incidents of vehicle vandalism. In the US, Waymo cars have been repeatedly damaged — slashed tires, smashed windows, traffic cones jammed on sensors. These stories aren’t anomalies anymore; they’re a pattern.

Online, the discourse has shifted too. Threads debating whether physical resistance against AI infrastructure is “justified” now surface regularly on Reddit and Hacker News. Most of it is thought experiment. But the fact that this conversation exists at all — and keeps growing — tells you something about where the public mood is heading.

The Luddites Weren’t Wrong

Here’s the part your high school textbook got wrong. The Luddites weren’t technophobes afraid of progress. They were skilled workers angry about how the gains were distributed. The machines created enormous productivity increases, but all the profit flowed to factory owners while workers got pink slips. Sound familiar?

MIT economist Daron Acemoglu has been making this point for years: the idea that technological progress automatically benefits everyone is a myth. Technology itself doesn’t cause backlash. The absence of institutions and policies to manage the transition does.

In 2026, the parallel is almost too neat. AI is sending corporate productivity through the roof. Stock prices soar. And the displaced workers get… a LinkedIn learning subscription and a severance package that covers two months of rent.

Violence Isn’t the Answer — But Neither Is Ignoring the Problem

Let’s be clear: violent resistance has almost never worked. The original Luddites were crushed by the British Army. The machines kept running. Smashing a data center won’t make GPT-5 disappear.

But dismissing the possibility of backlash is equally dangerous. Push automation without social safety nets and the reaction becomes unpredictable. Some security analysts are already warning that AI infrastructure could become a target for new forms of sabotage. Undersea cables, data centers, GPU supply chains — the physical chokepoints are more vulnerable than most people realize.

The real answer isn’t stopping technology. It’s building structures that share the cost of transition. Taxation on AI-driven profits. Large-scale reskilling programs that actually work — not the kind that produce a certificate and zero job prospects. Universal basic income experiments that go beyond pilot programs in Finnish towns. These conversations need to move from op-eds to legislation, and fast.

The Clock Is Ticking

South Korea offers an instructive preview. The country is adopting AI at breakneck speed, but its policy response remains stuck on “become an AI powerhouse” sloganeering. What makes Korea particularly interesting is its history of organized labor action — the massive worker struggles of the 1980s left a lasting infrastructure for collective mobilization. If large-scale AI displacement hits, the response there could be more organized and intense than anywhere else. That’s either a crisis or a catalyst for genuine social bargaining, depending on whether policymakers act before the pressure builds.

The same dynamic applies globally. Every government racing to lead in AI while neglecting the transition question is building the same pressure cooker.

When the Luddites picked up their hammers in 1811, what they really wanted to destroy wasn’t the machinery — it was an unfair system. In 2026, the same question confronts us: how do we distribute the wealth that AI creates? Answer too slowly, and history rhymes. The difference is that this time, nobody knows what replaces the hammer.