Tech giants are investing hundreds of billions of dollars into artificial intelligence infrastructure, with some estimates suggesting this spending could account for nearly half of the projected GDP growth in the U.S. this year. The influence of AI extends beyond macroeconomic figures, significantly shaping financial markets, particularly over recent months. Economics correspondent Paul Solman examines how AI is already transforming the economy and whether these rapid advancements could be fueling a speculative bubble.
Jerry Kaplan, a technology entrepreneur and academic, draws parallels between today’s AI race and earlier tech booms, noting that just as only a few dominant search engines or browsers survived past competition, only a handful of foundational AI models will likely endure. He predicts a market consolidation, with most current ventures failing to survive long-term.
Aswath Damodaran, a finance professor, cautions that expectations around profitability from AI products and services may be inflated. He compares the current enthusiasm to past overinvestments during the dot-com, PC, and social media eras, warning that revenue projections may not match reality. However, he also acknowledges that such overreach has historically driven innovation, arguing that progress often stems from bold, speculative ventures—even if they carry risk.
While the broader economy may remain resilient due to its size, the stock market’s heavy reliance on AI-driven valuations poses vulnerabilities. A sharp correction could erase trillions in market capitalization, particularly concentrated in a few major tech firms, potentially impacting retirement accounts and consumer spending. Paul Solman highlights concerns about job displacement, citing layoffs in tech sectors despite earlier demand for coding skills. At Stanford, even graduates with advanced computer science degrees are encountering difficulty securing employment.
Vivek Wadhwa emphasizes AI’s transformative potential, particularly in fields like medical diagnostics, where it could drastically reduce costs and improve access. Yet he also warns of widespread job losses in roles involving routine analysis or manual labor, alongside emerging threats such as deepfakes and cybersecurity risks. He stresses that the pace of change is so rapid that long-term outcomes remain uncertain—even without considering dystopian scenarios involving autonomous AI.
The discussion concludes with a reflection on the dual nature of technological progress: while it creates immense opportunities, it also brings disruption and unintended consequences.
— news from PBS
— News Original —
Is a bubble forming as AI investments drive economic growth?
Tech companies are pumping hundreds of billions of dollars into infrastructure to ramp up artificial intelligence. Some projections show that spending may be fueling nearly half of this year ‘s estimated GDP growth.
A.I. is having a disproportionate impact on the markets, too, especially in recent months.
Economics correspondent Paul Solman looks at how A.I. is already impacting the economy and whether a new risky bubble could be forming at the same time.
Jerry Kaplan:
There wasn ‘t a need for dozens and dozens of browsers or search engines. There ‘s only a need for a small number. We ‘re in exactly that same situation today with respect to A.I. There ‘s only a need for a handful of the basic low-level foundation models to require all this investment to build.
So what ‘s going to happen? Well, it ‘s all going to consolidate, obviously, and there will be only a few survivors left.
Aswath Damodaran:
There ‘s this dream, at least, that maybe one day we can make money from selling A.I. products and services.
Just as we overinvested for the Internet at the start of the dot-com boom, we overinvested in the P.C. architecture business before the P.C. boom, we overinvested in social media at the start of the social media boom, I think we ‘re overestimating how much money can be made from A.I. products and services.
Aswath Damodaran:
What ‘s so wrong with a bubble? They said, bubbles are terrible. They create pain. People lose money. I said, would you want to live in a world run by actuaries where it says, we ‘d still be in caves because they ‘d be looking at fire is a dangerous thing, let ‘s have a few more thousand years of testing things out?
Every advance in humankind has come from overreaching.
Paul Solman:
Yes, OK, but at what cost, if A.I. is, at the moment, driving economic growth overall, estimates as high as 50 percent of GDP this year?
So that, if the bubble bursts, it ‘ll be a real downturn for the U.S. economy.
Aswath Damodaran:
The real economy not that much, because if you think about the size of the real economy, even if you take all of the A.I. architecture investments, it is a small percentage of a huge economy, but the market, very different story.
Aswath Damodaran:
This is a market that has ridden on the back of A.I. for the last couple of years. The economy might do fine. Stocks might be down 40 percent. And that does have ripple effects in the economy.
I mean, losing $10 trillion in market cap just in five companies can have huge consequences for people ‘s 401(k)s, their spending.
Paul Solman:
That was CEO of the hot buy-now, pay-later firm Klarna this spring.
Ten years ago, I did a story on coding academies. It was the way to move up in the economy. And now profitable companies, very profitable, are laying off those very people, no?
Jerry Kaplan:
I ‘m actually seeing that in microcosm in my work at Stanford. A lot of the students are having considerable difficulty getting jobs. And I ‘m talking about the students graduating with master ‘s degrees in computer science. That ‘s a tremendous surprise for me. I haven ‘t seen that in my lifetime.
Vivek Wadhwa:
My A.I. can do what human beings can ‘t do in analyzing patterns. And, with it, I ‘m going to revolutionize the entire medical diagnostics industry. My goal is to be able to provide full comprehensive medical testing for less than a cost of a meal.
Vivek Wadhwa:
A.I. is going to take jobs on a scale that we haven ‘t imagined before. It ‘s not that all jobs will disappear. Many new jobs will appear. But a lot of existing jobs that do grunt work, that do manual analysis of stuff, that do manual labor will be eliminated because of what A.I. makes possible. This is unstoppable.
Vivek Wadhwa:
I talk about its good side, the fact that I can use it now to solve the problem of the medical diagnostics, to help reduce suffering. At the same time, it ‘s going to take jobs away and create all sorts of new nightmares, security issues, deepfakes, God knows what. None of us really understand how fast things are moving and where we ‘re going to be even five years from now.
Paul Solman:
And we ‘re not even talking about rogue A.I. that operates on its own, nightmare scenario of the “Terminator” movies, not talking about it in this story, anyway.
So, want a closing thought? Mine may well be the same as yours. The lord giveth and the lord taketh away.
For the “PBS News Hour,” Paul Solman.