路透报道 AI 正在降低小企业创业成本并引发美联储对生产率与就业影响的讨论

Here Now Health isn't an AI-based company, but its quick path from an idea in founder Michelle Turner's mind to an operating mental health platform for foster children relied deeply on the technology now restructuring the U.S. economy.

Working from her Virginia Beach home, she used AI tools to school herself in startup culture, create a business plan and fine-tune a ​presentation for early-stage investors. Funding began to flow, and the company, launched in January 2025, now has 16 employees and is certified in three states to provide Medicaid-funded mental health counseling for children entering the foster system, a gap ‌in care she identified through her own experience as a foster parent.

"A mom of six kids who's a first-time founder, who's a sole female founder, should not be able to raise (venture capital). I don't have an MBA. I don't have these things to back me up," Turner said. Developing her funding pitch with AI guidance was "like going to a master's level class every day from the robot. It was my startup advisor."

The rapid emergence and spread of AI has become a defining feature of the U.S. economy, of prime interest to Federal Reserve officials trying to understand its potential to reshape productivity, growth, inflation and labor demand. Among a broad review of the ​Fed launched by new Chairman Kevin Warsh, one panel will look solely at AI and its implications for productivity, a force that can allow the economy to grow faster with less inflation, but also means fewer workers are needed to create the same ​output.

Some Fed officials have already raised the possibility of an AI economy with structurally higher unemployment; other analysts have noted the steady dip in labor's share of national income and questioned if steadily higher returns ⁠to capital are also part of the AI future, a who-wins issue with social and political implications.

Much like when then-exotic new brands such as Yahoo! and America Online fought to connect everyone to the internet, different AI models are now competing for attention and dollars — but with ​offerings that can perform complex tasks, solve problems, or write computer code, not just shop and search the Web.

Investment in needed data centers is driving overall growth and in some cases driving up power and labor costs. Scenarios around AI range from visions of plenty to mass unemployment, with ​banks, government agencies, the military and more looking to exploit the new tools but also protect from them.

“Markets are confronted with dramatically different competing narratives," Jean Boivin, head of the BlackRock Investment Institute, said in a seminar with journalists on Tuesday. "We are framing this as scarcity versus abundance…Scarcity is the story of the moment” with the AI investment boom driving up some costs and driving demand for capital, but, "we are also talking about abundance….We are talking about AI that can lead to significant breakthroughs…Growth that might be breaking out of a 2% world.”

Somewhere in the middle sits Turner's journey from non-profit manager to chief executive of a growing company, a ​story John Bailey, a nonresident senior fellow at the American Enterprise Institute and adviser to one of the firms that invested in Here Now Health, says is becoming more common.

For small entrepreneurs "things that used to take too much time or cost too much — the price to access has ​fallen close to zero," said Bailey, who helped Turner from the outset develop the AI tools she relied on. "It is empowering entrepreneurs to scale faster and hire people. These are not AI companies. They are traditional companies trying to deliver services but do it faster, cheaper."

Public debate remains focused on AI's job-disrupting ‌capacity, with rounds ⁠of tech industry layoffs attributed to it and evidence it is letting firms reduce back-office and clerical employment. Bailey said he's become more convinced the AI economy won't kill jobs so much as change and rearrange them, much as happened with earlier technological leaps.

He isn't alone in seeing upside.

Torsten Slok, chief economist at investment firm Apollo Global Management, attributes a recent upturn in already-high new business formations to AI, which is "dramatically reducing the cost and complexity of launching a company. As these firms scale, they will create jobs."

It may take years to know how it nets out.

A recent jump in job creation has eased concern that the U.S. was already entering an era when technology would replace labor at such a pace that unemployment might rise and mark the first time that new general technology proved, on net, labor destroying.

Richmond Fed President Thomas Barkin said in an ​interview earlier this year he is wrestling with the employment risks AI ​might pose, but also struck by firms telling him it ⁠is easing worker shortages in some skilled occupations, a boost to supply.

"We are all quick to see the disasters, which is about jobs getting replaced," Barkin said. But he said contacts in areas like auto repair or manufacturing generally "are still in a world of saying they cannot get enough workers," and are leaning on AI to make whoever they can hire more productive.

"It is still going to be a challenge. It is a 'rust-belt ​risk'," for some white-collar occupations in particular, Barkin said. But "we are not an economy that has no shortages."

The transition, however, may be painful.

The globalization of world trade in the 1990s decimated longstanding U.S. ​manufacturing clusters. Programs to transition workers to ⁠new jobs were not effective. Over time the narrowed opportunities in parts of the Midwest and South are felt to have contributed to a rightward drift in politics and "deaths of despair" from substance abuse.

Researchers in a recent study warned a similar shock may be developing for clerical, administrative or other workers, particularly those without college degrees who rely more on work experience to advance their careers. The Brookings Institution and Opportunity@Work study found about 23 million people whose logical next step in a career was into a job highly exposed to AI replacement - in effect putting them at risk of being stranded in lower-paying ⁠roles.

"Disruptions in these roles ​can have outsized effects on workers’ ability to move into higher-wage work," the researchers wrote, with regional impacts focused in Florida, the Northeast, Texas and California, hubs of ​the sort of work susceptible to AI disruption and a different geographic spread from the earlier manufacturing disruption.

For the Fed, both the outcome and the pace of the AI transition will be important, with short-term impacts potentially very different from those seen in the long run, when it will become clearer if a productivity boom occurs.

At his debut press ​conference, Warsh called AI the most important economic change "that we've had in my adult lifetime," and said the U.S. "is ultimately going to be better off" because of it.

But "that certainly doesn't mean it's not going to be disruptive," he said.