In conversations with hundreds of business leaders over the past few months, I’ve seen a sharp divide in their views of artificial intelligence. One camp sees a “job apocalypse” and mass unemployment ahead; the other sees a great leap forward for society.
Put me in the second camp — with a few caveats. Will A.I. disrupt the labor market? Absolutely. This transition, like other significant moments in our history, will entail new challenges, especially as A.I. separates labor from productivity in magnitudes we haven’t seen before. But the United States has a long track record of creating new jobs in response to disruption, from the electrification of the 1900s to the digital revolution of the 1990s; I don’t see any reason to think this dynamic will stop now.
There’s no question A.I. will reshape our everyday lives. Goldman Sachs’s economists estimate that, over the next decade, A.I. may automate 25 percent of current work hours. While it’s difficult to see how people in hands-on professions like food preparation, construction or services will be affected, people in white-collar jobs, among them accountants, bankers and lawyers, will likely see many of their tasks automated. According to one Stanford study, in the occupations most susceptible to greater automation, such as software engineering or customer service, entry-level employment has already declined by 16 percent relative to the least-exposed roles.
But when you look at jobs or sectors less relevant to automation, the picture changes. Our economists estimate that the growing demand for data centers has created more than 200,000 construction jobs since 2022. While A.I. eliminates jobs in some sectors, it may lead to job growth in others. Goldman Sachs may need fewer people in regulatory reporting or client onboarding, freeing us up to hire more bankers, traders and asset managers who are interacting with clients constantly.
Of course, we can’t dismiss the human cost of such disruption. The Industrial Revolution raised living standards only after society endured the hardships of grueling labor in mills and mines and the fetid slums that came with rapid urbanization. In recent decades, manufacturing employment has declined significantly owing to automation and global outsourcing. This caused enormous hardship for many families and communities across America such as Gary, Ind., and Greenville, S.C.
But for all those challenges, I keep bumping up against this reality: Standards of living for a vast majority of Americans are significantly higher than they used to be. When I was born in 1962, the average American adult didn’t have air-conditioning, but as air-conditioner prices dropped, nearly all of us got cool. In the 1950s, only a few large corporations, like IBM, had computers; now some 90 percent of American adults walk around with a supercomputer in their hand. In 1900, global life expectancy at birth was 32 years old; today, it’s over 70.
Perhaps more to the point, job growth has outpaced population growth. Since 1962, civilian employment in the United States has increased by roughly 145 percent, while the civilian population age 16 and older has risen by about 128 percent. In that time, we’ve seen new sectors emerge as others have grown or faded. While manufacturing employment declined from 15.5 million to 12.5 million over this period, led by almost two million jobs lost in textile and apparel making, the health care industry now employs more than 18 million workers. The U.S. economy is still the most innovative, dynamic and entrepreneurial in the world.
It’s true that even the most reliable historical patterns can be broken, but there are three reasons I expect the U.S. economy to remain as resilient and dynamic as ever.
First, if our estimate proves correct, A.I. won’t eliminate 25 percent of jobs. What’s more likely is that people will find more productive ways to spend their time. When I was a first-year banking analyst, something as simple as making a graph of a stock’s performance took six hours of looking up prices in back issues of The Wall Street Journal on microfiche. Today, a first-year analyst can do it in seconds, and we have employed more people than ever in recent years. With more sophisticated tools, the complexity of our work naturally expands. Do any of us feel like we have less to do these days despite the convenience of Excel, email or Zoom?
Second, just because a job can be replaced doesn’t mean it will be. The television didn’t eliminate the demand for live entertainment, nor did the internet put real estate agents or fitness instructors out of business. Instead, these technologies highlighted and enhanced the value of those professions. Technological change and cultural change do not move in tandem. After all, even after decades of A.T.M.s, digital banking and bank consolidation, employment in commercial banking is at roughly the same level today as it was in the mid-1990s.
Third, the American labor market is vibrant. While net job creation is a few million per year at most, the gross number is much higher; American companies destroy and create between 25 million and 35 million jobs annually. It’s easy to imagine the pace will ramp up as A.I. enables even more innovation, and we’re already seeing the economy adapt. Companies are now looking for people to manage so-called agentic A.I. and incorporate it into a wide variety of areas, from implementation and workflows to compliance and validation. All of this requires human judgment.
If A.I. does indeed destroy jobs — and at a potentially greater speed than we’ve seen — then public policy must respond, by funding large-scale reskilling or encouraging A.I. that supports workers instead of replacing them.
This must be a joint effort between the public and private sectors. The public sector should create incentives and provide resources where necessary, including investing more in vocational schools and community colleges; the private sector should help workers upgrade their skills and reimagine on-the-job training.
The historical pattern is clear: The U.S. economy can and will adapt to major advances in technology. What’s also clear is that stark forecasts by even the most brilliant minds often miss their mark. In 1930, John Maynard Keynes famously predicted that, by 2030, people would work only 15 hours a week. While his vision of a leisure-filled future remains unfulfilled, it is a good reminder that fears of a job apocalypse may very well overlook A.I.’s potential to spur an economic and productivity revival.
David M. Solomon, in addition to running Goldman Sachs, is an electronic dance music producer known as DJ D-Sol.
The Times is committed to publishing a diversity of letters to the editor. We’d like to hear what you think about this or any of our articles. Here are some tips. And here’s our email: [email protected].
Follow the New York Times Opinion section on Facebook, Instagram, TikTok, Bluesky, WhatsApp and Threads.
The post I’m the C.E.O. of Goldman Sachs. The A.I. Job Apocalypse Is Overblown. appeared first on New York Times.
