Ownership Over Optimism: Rethinking AI Wealth for Workers
Ramaswamy says workers can build wealth in the AI era. Here’s the thing: saying workers can get rich off better tools is like saying farmers got rich after the reaper was invented — possible, but mostly for the ones who also owned the land and the storage and the rail connection to market.
The headline promises a playbook. The logic is familiar: technology raises productivity; productivity should raise incomes; therefore workers can capture value. I don’t really quarrel with the first two steps. AI is clearly adding productive capacity — from code to copy to customer service. Yeah, no one in Silicon Valley seriously doubts that anymore; we’re past the “is this real?” phase and deep into “how do we ship this before Q3?” territory.
But productivity gains don’t magically show up on household balance sheets. They show up as profits unless you build very specific pipes that route some of that flow to workers.
Ownership, not just “more skills,” is the fulcrum
Look: we talk about “reskilling” like it’s WD-40 for the labor market. Squeak here, spray a course there, problem solved. Learning to prompt a large language model or automate a routine task does make an individual more employable. It does not, by itself, create equity.
If a warehouse worker learns to operate an AI-driven sorting system but still earns a fixed hourly wage, the company captures the surplus. That’s not a morality tale; it’s just the way returns behave in a system where capital owns the tools and the data. As long as AI systems sit firmly on the asset side of the corporate ledger, the default is that gains flow to shareholders.
History already ran this experiment. Industrial mechanization in the 19th century sent output through the roof, but the early gains mostly accrued to factory owners until unions, regulation, and new ownership vehicles forced a re-split of the pie. You can practically hear William Gibson cackling from Neuromancer: we keep building better tools and acting surprised when the folks who own the tools write the ledger.
If the goal is AI as a driver of broad-based wealth, the boring implementation details matter a lot: employee equity, profit-sharing, tax policy that doesn’t punish shared ownership, and benefits that follow workers across gig, contract, and full-time roles. These are plumbing decisions, not TED Talk lines.
Employee stock plans and ESOP-like structures are the obvious reference point, but they’re unevenly used and heavily concentrated in certain sectors. Tech startups live on options; a regional logistics company or a family-owned restaurant usually doesn’t. Any serious claim that “workers can build wealth” has to confront that not all firms have the same access to capital, margins, or tolerance for sharing upside — and that wealth-building plays out over years, not the length of a bootcamp.
One historical wrinkle that rarely gets mentioned: there are examples where spreading ownership changed outcomes at scale. Think of broad-based stock plans at companies that made regular, predictable grants part of compensation instead of lottery tickets tied to a one-time IPO. The lesson isn’t that everyone got fabulously rich; it’s that ownership turned abstract productivity booms into actual household assets for more than just founders and investors.
Skills are necessary but badly oversold
AI will absolutely mint new roles and winners. Some workers will ride that wave into real wealth: creators who figure out how to monetize AI-produced content, engineers who start companies, consultants who package AI into high-margin services. Sure, but that’s the head of the distribution, not the whole curve.
I’ll be honest: the narrative of mass uplift through retraining alone feels like half economics, half self-help genre. Real retraining takes time, money, childcare, and sheer luck that local employers actually need the new skills being taught. Plenty of workers finish programs only to discover the “hot job” exists two states away or mostly inside firms that don’t hire from their community college.
There’s also the geographic skew. Urban tech hubs and big metros will grab a disproportionate share of AI-heavy roles. Legacy industrial towns and rural regions will see slower adoption, thinner job markets, and fewer ladders from “learning the tools” to “owning a piece of the action.” Think of it as the inverted world of Neal Stephenson’s The Diamond Age: the nanotech primer exists, sure, but most people never get it handed to them.
The standard rebuttal is easy to recite: AI democratizes capabilities. A solo designer with great AI tools can compete with a big agency. A local retailer can run targeted marketing that once required a whole department. That’s real, and we’ve already seen micro-entrepreneurs turn cheap software into respectable income.
But pockets of success aren’t the same as a generalized path to wealth. Platform economics have a bad habit of funneling most rewards to a sliver of participants. The app store, the creator economy, ride-hailing drivers — we’ve run this movie. New capabilities expand the opportunity set, then a handful of platforms and top performers absorb most of the winnings.
If you want wealth, you need a claim on the upside
If policy is going to tilt outcomes, it has to go beyond writing checks for training programs and calling it a day. Incentives for employee ownership, tax treatment that doesn’t make broad-based stock a nightmare, easier vehicles for turning freelance income into long-term savings — those are the mechanisms that convert AI productivity into actual household assets instead of just nicer quarterly earnings calls.
Companies aren’t helpless here, either. They can push broad equity grants further down the org chart. They can build revenue-sharing for platform contributors instead of just one-time bounties. They can design apprenticeship tracks where learning the tools comes bundled with a literal stake in the products those tools help create. It’s not charity; it’s a different way of structuring incentive and loyalty in a world where skilled labor has more outside options.
There is real political and corporate risk baked into all of this. Anyone arguing for widespread sharing of corporate returns is going to collide with investors and executives who see their slice as the logical payoff for taking risk. That tension isn’t going away; it’s the main event. Any upbeat story about AI and worker wealth that ignores it is writing fan fiction, not a playbook.
So what does an individual worker actually do with all this? Learn the AI tools, obviously. But also start treating ownership as non-optional: push for equity or profit-sharing when you can, steer your career toward firms and roles where upside is even theoretically on the table, and treat “what’s my stake in the value this creates?” as a standard question, not an awkward one.
Silicon Valley lives and dies by options and cap tables; most Americans still negotiate only on salary and healthcare. If AI really is the next big productivity wave, the workers who translate between those two languages earliest are going to look unreasonably prescient a decade from now.