Questioning Ramaswamy's AI Wealth Promise for Workers

The Wall Street Journal column says workers can build wealth in the AI era. A hearty cheer for optimism. But optimism doesn't pay for training, child care, or the months a displaced worker lives between paychecks.

Let’s grant the appeal first.
Vivek Ramaswamy’s message — that workers should pivot, capture upside, and become stakeholders in the technologies reshaping work — has moral force. Who wouldn’t applaud people owning more of their gains? The aspiration is right.

The architecture is wrong.

Point one: the prescription ignores where capital actually moves.
Ownership opportunities don’t land on a level playing field. Venture capital, stock options, and private-equity exits concentrate in specific geographies and established firms. Follow the money. The financing that seeds startups and funds employee equity is cyclical, network-driven, and often demands years of tolerance for volatility — a luxury not available to hourly retail workers, bus drivers, or the millions in industries AI will hollow out first.

If wealth-building depends on equity-like instruments, then access to them matters more than encouragement. A subscription to an online course isn’t the same as a board seat or meaningful equity. Suggesting that workers can “build wealth” without addressing the machinery of capital allocation is like telling people to bake a cake without giving them an oven.

History backs this up.
When tech waves hit — think the early internet boom — the people who captured most of the upside weren’t the ones dutifully “reskilling” in night classes. They were early employees with stock, founders with backers, and investors who could afford to lose money for a while. Workers who joined later, even at the same companies, often got salaries and slogans, not transformative ownership.

Point two: incentives inside firms pull in the opposite direction.
The column leans on the idea that employers and workers will share the upside of AI. Convenient, isn’t it. But corporate governance and incentive design rarely default toward broad-based employee ownership. Executives answer to shareholders and boards; investors push for scale and margins. When automation increases productivity, firms often prioritize returns to capital — dividends, stock buybacks, executive compensation — over wide employee participation.

Ask who designs compensation plans. Ask who sits on compensation committees. Workers rarely hold decisive sway there. Policy nudges can change that — tax credits for employee stock plans, legal frameworks for employee-owned businesses — but the column treats these shifts as cultural choices rather than institutional rewiring. Culture is the press release; bylaws are the power.

You don’t have to look far for a cautionary tale. When companies roll out AI tools to replace back-office staff or call-center workers, the savings often get framed as “efficiency.” They show up first in earnings calls and share prices, not in line workers’ 401(k) statements. The story of automation has usually been: gains to capital now, promises to labor later.

Point three: retraining is necessary but not sufficient.
The article argues, implicitly, that upskilling and entrepreneurship are routes to wealth. Yes. But retraining programs have a record of mixed results when they’re disconnected from employer demand, child-care support, transportation, and a realistic timeline for income replacement. Telling a laid-off worker to “learn to code” or launch a micro-business ignores the months of unpaid grind between enthusiasm and a single dollar of profit.

Entrepreneurship carries risk. The myth of the “side hustle to riches” glosses over capital constraints, market access, and the often invisible administrative burdens that sink small ventures: licensing, insurance, compliance, platform fees, and the constant scramble for customers. Here’s what they won’t tell you: for every founder who wins big, there are many who absorb the losses quietly.

Now, the hard counter-argument to my critique: maybe market forces really will create abundant, low-friction pathways — AI tools that democratize entrepreneurship, gig platforms that enable new income streams, remote work that widens opportunity. That’s plausible. Tech does lower barriers in some cases.

But plausible is not inevitable.
Without deliberate policy and institutional redesign, those same market forces still favor repeatable scale and incumbent advantage. Platforms can concentrate power; AI models can be expensive to deploy; discoverability on crowded marketplaces is scarce; contract work often lacks benefits. The result: more people trying, fewer succeeding, and a widening gap between the headline success stories and the median worker’s reality.

Two steps the column skips but should have foregrounded.

First, democratize ownership structurally. If the argument is that workers should share in AI’s gains, then propose concrete mechanisms: mandatory profit-sharing options for firms above a certain size; incentives for broad-based employee stock ownership plans; seed funds targeted at worker cooperatives adopting AI to boost productivity rather than replace labor. Don’t just tell workers to be entrepreneurs — change the rules that make entrepreneurship survivable.

Second, recognize sectoral divergence. AI will uplift some professions and decimate others. A one-size-fits-all pep talk won’t do. Public policy can focus resources where displacement risk is highest: transition apprenticeships, portable benefits, wage insurance, and support for regions that become collateral damage in someone else’s efficiency story. Yes, these are costly. Follow the money — the cost of doing nothing will be paid in foregone consumer demand and social strain.

Concede this: the column’s optimism has rhetorical value. Encouraging agency matters. People shouldn’t be told they have no options.

The question isn’t whether workers can, in theory, build wealth in the AI era, but whether the ownership architecture shifts fast enough to make that column’s promise more than a shareholder’s fairy tale.