Rethinking AI Wealth: Workers Need More Than Hype

Ramaswamy says workers can build wealth in the AI era. Noble claim. Plausible in a thought experiment. Dangerous if treated as a strategy.

Because which workers? What wealth?

Those questions matter, and not just in a seminar-room way. His Wall Street Journal column hangs on a premise that can be true for a few and ruinous for many. Follow the money.

Let’s start with what he gets right. Workers should chase upside, not just cling to jobs. Skills, ownership, entrepreneurship—these are not villainous ideas. They’re survival tools in an economy that rewrites job descriptions faster than HR can update the portal.

But the promise that “workers can build wealth in the AI era” dates itself to whoever owns the tools.

The pitch—workers will adapt, seize new opportunities, capture gains—is the oldest act in the tech playbook. Learn new skills. Start a company. Buy equity. Sounds tidy. Who actually gets first crack at those prospects?

Ownership matters more than skill when capital is concentrated. If AI systems are developed, deployed, and monetized by a handful of firms, the gains compound inside corporate profit and investment returns. Workers might see a temporary spike in demand for certain tasks; they won’t automatically inherit the patent, the platform, or the distribution channel that turns automation into real wealth. Why would they? Companies answer to shareholders and boards; employees answer to paychecks and rents.

Training is only half the equation. Access to capital, networks, and time is the other. A gig worker can learn prompt engineering between shifts, sure. But that worker rarely gets equity in the platform that replaces or “uplifts” them. And if the new “wealth” is largely paper—stock options in companies only a venture capitalist can exit from—what looks like democratization is often a rebranding of familiar venture gains for the already connected. Convenient, isn't it.

Here’s what they won’t tell you: AI-driven productivity can lower wages as easily as it can raise them. Firms can pocket the gains as higher margins instead of higher pay. Without institutional change—bargained wages, profit-sharing, stronger antitrust enforcement—the distribution defaults to capital.

History has already run this experiment. Think about the last boom built on “everyone’s an owner”: the ride-hailing era. Uber and Lyft told drivers they were entrepreneurs, not employees. A few early staffers and investors walked away with fortunes. Many drivers walked away with a depreciated car and no safety net. The story wasn’t about who hustled hardest. It was about who wrote the terms of ownership.

Ramaswamy’s optimism assumes broad-based participation in that ownership. That’s good rhetoric and thin scaffolding.

There are real routes to shared upside: worker cooperatives, employee stock-ownership plans, public stakes in AI infrastructure. Each demands deliberate choices by policy-makers and corporate leaders, plus the bruising work of politics—over tax treatment, over who finances national AI compute, over whether critical tools are treated like utilities or locked-up assets. None of that happens because someone wrote an upbeat op-ed.

Don’t expect CEOs to volunteer away their power. They sell visions; investors buy growth. Little in corporate law or incentive structures compels management to distribute windfalls to rank-and-file workers. Say the CEO sounds magnanimous in a guest essay. Who appoints the compensation committee? Who ties management bonuses to profit-sharing formulas instead of stock price alone? Follow the money.

Geography deepens the fault lines. High-tech hubs concentrate not just engineers but venture capital, specialized law firms, and the informal networks that get you into the next deal. Rural and older industrial communities see the mirror image: thin capital, shallow networks, limited bargaining power. The idea that training alone will equalize outcomes is airy unless you also reshape where capital flows and who controls the platforms.

Here’s another blind spot. The column reads like a story of linear progress: AI arrives, workers adapt, wealth spreads. But technological shocks aren’t neat. The first wave of power looms crushed artisanal weavers before new factory jobs stabilized anything. The early internet enriched a band of platform builders while hollowing out local newspapers and retailers. Workers didn’t fail to “upskill” their way out of those transitions; the benefits were designed to pool elsewhere.

Critics of this critique will invoke the usual script: innovation has historically created new industries and jobs; today’s displaced worker becomes tomorrow’s entrepreneur or specialist. That’s true in many cases. But past technological waves were cushioned, however imperfectly, by unions with real bargaining power, public education expansions, and an economic compact that tied productivity more closely to wages. Those aren’t givens now. The AI wave can move faster than the institutions that would soften it.

If you want Ramaswamy’s thesis to be more than a pep talk, you need mechanisms, not mantras.

Make employee ownership real by changing how worker equity is taxed and regulated so it isn’t a perk reserved for white-collar staff at marquee firms. Build regional AI training and deployment centers that are tied to local employers and public agencies, not just elite universities. Use public purchasing power—state, municipal, federal—to favor AI deployments that upgrade jobs rather than erase them.

And while we’re talking about ownership, look at who owns the models and data. When OpenAI or Google rolls out a new tool trained on the world’s text and images, the line between “public input” and “private profit engine” gets thin. Workers already contributed value—through their writing, coding, art, and feedback—long before they’re told to “adapt” to the output. That asymmetry is part of this story too.

Ramaswamy’s column operates as a manifesto of possibility. Possibility is not policy. You can cheerlead adaptation all you want; it doesn’t rewrite corporate charters or shareholder agreements.

Workers can build wealth in the AI era—but only if the same creativity being lavished on the technology is aimed at rewiring who owns the gains. Otherwise, the slogan will age nicely on a conference panel while the wealth quietly pools where it always has.