Why AI Won't Deliver Worker Wealth, Not Today

Ramaswamy says workers can build wealth in the AI era. Look — that’s not nonsense. Technology absolutely creates new value, and some workers will ride that wave.

But the Wall Street Journal argument treats “workers” like a single, mobilized army with equal access to tools, capital, and opportunity. That’s never been the case. It won’t be now.

I ran operations at a Fortune 500. I lived inside headcount spreadsheets, training budgets, and performance stacks where “potential” met quarterly earnings. I’ve seen well-intended “upskilling” programs die in committee once someone asked, “Where’s the ROI?” So I’m sympathetic to optimism about individual initiative. I’m just not willing to confuse it with a plan.

Nice slogan. Weak scaffolding.

Here’s what nobody tells you: telling workers they can build wealth is easy. Giving them a reliable path to do it is not.

The piece is right about one thing — technology creates new value that can, in theory, be shared. Historically, new tech waves have done that: factory workers who stuck through early industrialization, office workers who moved into computer-enabled roles, logistics teams that adapted to global supply chains. Gains were real.

But they weren’t automatic, and they weren’t evenly distributed.

The argument here assumes workers can capture AI-driven value through choice and hustle alone. It skips the scaffolding questions that decide who actually gets ahead: Who pays for meaningful reskilling when the old role disappears? Who front-loads capital when displaced workers need a bridge, not just a webinar? Who makes sure ownership opportunities don’t stop at founders, investors, and a thin corporate leadership tier?

Give me a break — “take initiative” doesn’t pay for a six-month gap between jobs or replace lost benefits.

A lot of op-eds read like pep talks for individuals. Real change lives in the plumbing: compensation design, benefit portability, training budgets, capital allocation, and the power to say no to bad offers. Those are operational tools I’ve seen pulled, tested, and undermined once the bill came due or leadership changed.

The distribution problem — not just a training problem

Ramaswamy’s claim is plausible if you imagine a world where training is free, mobility is frictionless, and every worker has equal access to financial tools and networks. That world does not exist.

The more consequential issue isn’t whether AI creates wealth; it’s who gets first claim on it. AI tends to reward whoever owns the models, the data, and the infrastructure they run on. Workers in roles that complement those systems can see gains. Workers whose tasks can be decomposed into prompts and workflows don’t just lose tasks; they lose bargaining power, and often the ability to walk away.

Regions and industries split along those lines. A software engineer in a major tech hub has a different opportunity set than a call-center worker three layers down a subcontracting chain. Same “AI era,” wildly different balance sheets.

This is a distribution problem more than a technical one. Policy and corporate design choices — profit-sharing, equity for rank-and-file, portable benefits, publicly funded re-skilling — determine whether the gains diffuse or concentrate. Saying “workers can build wealth” without naming those mechanisms is like recommending a house without explaining the foundation.

We’ve seen this movie. When railroads scaled, railroad owners and financiers made fortunes long before line workers saw stability. When the internet took off, broad stock ownership was talked up, but concentrated cap tables and stock-based pay tilted heavily toward executives. AI will replay that pattern unless the underlying rules change.

What actually helps workers — and what’s theater

Here’s what nobody tells you: the interventions that work are boring and often politically unsexy.

Profit-sharing tied to simple, auditable metrics. Mandatory notice periods plus wage insurance for displaced workers so the “retraining” window isn’t just a euphemism for panic. Tax incentives that reward companies for giving non-executive equity and for funding portable training accounts. Apprenticeship programs co-designed with employers and community colleges that lead to specific roles, not just certificates.

These tools change incentives and make wealth-building feasible for ordinary workers. Not more inspirational op-eds. Not panels about “lifelong learning” that never touch how people pay rent while they learn.

There’s a difference between worker-facing theater and worker-facing power. Theater is another AI webinar, a self-paced training portal, and a talking point in the annual report. Power is the ability to own part of the upside, to keep benefits while you move between gigs, and to say no to exploitative terms because you’re not one paycheck from disaster.

Design work, don’t just “train” workers

Employers have more control than they admit.

They can redesign roles so the human contribution is clearly complementary to AI — judgment, relationship-building, domain expertise — instead of turning every worker into a thin layer of oversight on top of automated workflows. That’s how you avoid turning mid-career professionals into low-paid “prompt operators” with no bargaining power.

Cities and states can pilot portable benefit accounts so a worker doesn’t lose retirement or health stability when they switch employers or move between W-2 and 1099 work. This isn’t some utopian idea — it’s just treating workers like long-term assets in the economic system instead of disposable line items tied to one payroll.

Those changes are tactical, measurable, and scalable. They require buyers, executives, and lawmakers to stop treating labor policy as a moral plea and start treating it as operations improvement and risk management.

Markets help — but not on their own

You’ll hear the counterpoint: markets will sort this out; innovation will create new jobs; entrepreneurs will democratize ownership.

Parts of that are true. Markets do create opportunities. You see it when small AI consultancies, niche product builders, and solo operators carve out value by combining tools in new ways. Some workers will absolutely use those channels to build wealth.

But markets also concentrate returns when barriers to capital, information, and networks persist. If you’re already well-capitalized, well-networked, and sitting close to the AI supply chain, you catch the upside early. If you’re not, “opportunity” often shows up as instability and churn before it ever looks like wealth.

So act.

Firms can pilot profit-sharing and portable training accounts now, not after another layoff wave. Policymakers can stop treating worker wealth as a happy accident of growth and start writing rules that assume AI’s gains will otherwise pool at the top. Workers can demand concrete terms — equity plans, severance norms, retraining guarantees — not just slogans about “opportunity in the AI era.”

Bottom line: Ramaswamy’s promise that workers can build wealth under AI will come true for a narrow slice by default; whether it scales beyond that depends on who’s willing to fight over the plumbing, not the prose.