FW: Nw York Times: The Morning: An A.I. bubble?

[neillh...@earthlink.net]

 

 

From: Mark Woodall <woodal...@gmail.com>
Sent: Monday, October 27, 2025 8:00 AM
Subject: Nw York Times: The Morning: An A.I. bubble?

 

 

---------- Forwarded message ---------
 

 

Plus, President Trump’s Asia tour and Hurricane Melissa.

View in browser|nytimes.com   October 27, 2025 Good morning, and happy Monday. Here’s what’s happening today: Asia tour: President Trump is in Japan. He met with the emperor and is scheduled to meet the new prime minister tomorrow. Trump ally: Javier Milei, Argentina’s budget-slashing president, won big in midterm elections. It’s a sign that many voters still back his austerity policies. Russia: Vladimir Putin said his country now had a nuclear-powered missile. We look at Hurricane Melissa and America’s Halloween candy choices. But first: Is A.I. a bubble?   Brendan McDermid/Reuters Reality check By Evan Gorelick I’m a writer for The Morning. How well is the stock market doing? That now depends, almost entirely, on artificial intelligence and the companies that make it. Those companies, concentrated in Silicon Valley, are spending hundreds of billions of dollars. Investors think this tech is the future, so they’re snapping up shares. By one estimate, 80 percent of U.S. stock gains this year came from A.I. companies. Does it feel like a bubble? Big names like Jeff Bezos (Amazon), Sam Altman (OpenAI), Jamie Dimon (JPMorgan) and David Solomon (Goldman Sachs) worry we’re on the cusp of a correction. They warn that valuations are getting too high and that eventually, reality may bite. And if those companies plunge, they’ll take the economy with them. But nobody is acting like we’re in trouble. The S&P 500 has notched more than 30 record highs this year. The wisdom of the bet rides on a basic question: Is A.I. a boom or a bubble? Here are the arguments. A boom If A.I. tech can keep getting better, then A.I. companies can keep getting more valuable. Many believe it. These companies assume that building out A.I. infrastructure — more chips, more data, more power — improves the technology coming out the other end. This is called the “scaling hypothesis.” It helps explain why Nvidia and AMD recently announced $100 billion deals to install their chips in OpenAI’s data centers; why Amazon is spending more than $100 billion on A.I. data centers this year; why Meta will spend more than $600 billion over the next three years; and so on. Sales at Nvidia jumped 56 percent last quarter. That could mean more growth ahead, justifying the dizzying share prices and massive investments. My colleague Joe Rennison, who covers financial markets, sums it up nicely: Valuations have risen very quickly, but that’s because the main companies like Nvidia have had the earnings to back them up. We haven’t sated demand for chips yet, so we don’t know what the ceiling is. Steady spending has allayed worries on Wall Street. Companies haven’t lost faith in A.I., so investors haven’t either. A bubble But what if A.I. doesn’t quickly transform the economy? Then, businesses and users won’t grasp as eagerly for it. That could spell ruin for investors who’ve staked their financial futures on an A.I. revolution. Seven tech companies now account for more than a third of the value of the S&P 500 and trade at prices 70 times higher than their earnings, on average. And while nearly eight in 10 businesses already use generative A.I., just as many report “no significant bottom-line impact” from doing so. The spending spree, explains my colleague Cade Metz, who covers A.I., is based on faith: They’re banking on those scaling laws continuing indefinitely. They’re assuming that these really smart people at these companies will keep coming up with new ideas. These investments, in other words, are speculative. Companies like Oracle are taking on billions in debt under the expectation that the tech will advance, but there’s no guarantee. OpenAI is still losing money, and many A.I. companies continue to buy from and invest in each other in “circular” deals, my colleague Andrew Ross Sorkin wrote this month. If it becomes clear, at any point, that A.I. will fall short of expectations, then investors will pull back, and the bubble will pop. The stakes We’ve been here before. In the late 1990s, investors poured their money into internet-based companies. Telecom firms built out infrastructure — tens of millions of miles of fiber-optic cables — to support the anticipated boom. It was a bubble. The profits never materialized, so investors sold off their shares and the companies collapsed. The Nasdaq lost more than three-fourths of its value, and it took 15 years to recover. This time, the companies involved are much bigger. Yet unlike their dot-com forerunners, they’re actually earning some money. Just not as much — for now — as investors hope. More on A.I.: Big tech companies are making the Cal State college system a training ground for A.I. tools in education.