Everyone's been fixated on generative AI coming for knowledge workers. But the bigger shake-up might be happening in the physical world, where robots are quietly eyeing jobs in warehouses, factories and delivery trucks.

Oxford Economics senior economist Nico Palesch told MarketDash that roughly 20% of the US workforce could face high exposure to physical robotics and automation within the next decade or two. That's not a small number—we're talking about one in five American workers.

Half of Transport and Manufacturing Jobs on the Chopping Block

The industries most vulnerable aren't in Silicon Valley. Transport and logistics leads the pack, with more than half of all positions at risk from self-driving technology and warehouse automation. Manufacturing isn't far behind at 51.1% exposure.

Accommodation and catering clocks in at 47.2%, retail at 40.2%, wholesale at 31.0%, and extraction at 35.1%. The pattern is clear: if your job involves moving things or making things, the robots are coming.

But here's where it gets uncomfortable. "The people losing their jobs aren't necessarily going to get new jobs," Palesch said bluntly.

Creative Destruction or Just Destruction?

When asked whether this wave of automation represents the kind of creative destruction economist Joseph Schumpeter wrote about—where old jobs die but new ones emerge—Palesch didn't offer easy reassurance.

Sure, the economy as a whole might adjust over time. Cost savings and productivity gains eventually fuel new investment and create jobs elsewhere. "We still haven't seen any evidence that we're going to need to see massive amounts of less human labor over the coming decades," he noted.

But that's cold comfort if you're a 50-year-old truck driver whose livelihood gets eliminated by autonomous vehicle software. You're probably not going to find equally skilled, equally paid work elsewhere. "For them unemployment will be quite a bit higher probably for many many years, potentially for the rest of their lives," Palesch said.

Even when aggregate economic statistics look healthy, individual communities can feel the political and economic fallout for generations. And structurally? "There can certainly be a structurally higher unemployment rate for the broader economy," he acknowledged.

Who Gets Rich When Robots Do the Work?

The trickiest question might be inequality. If capital increasingly replaces labor, the gains flow to whoever owns the robots and the AI—widening the gap between people who own assets and people who sell their time.

Palesch doesn't see that outcome as inevitable, though. AI could democratize opportunity, he argued, enabling ordinary people to "do their own businesses, to have their own ideas" and operate independently without needing corporate infrastructure.

As for universal basic income? Only viable in an "extreme scenario" where AI delivers a massive productivity windfall that expands the tax base enough to fund it, while simultaneously driving inequality so high that it becomes politically necessary. Given current government debt levels, that's a pretty narrow path. Absent a genuine productivity revolution, the math just doesn't work.