AI stocks have been hammered, but data centers are still expanding at a furious pace! Has Goldman Sachs spotted an opportunity others have missed?

The AI market has been a little surreal lately: tech stocks have fallen so hard that investors are questioning everything, while the data center industry is still preparing to spend more.
Goldman Sachs raised its forecast for U.S. data center construction spending, projecting it will reach $125 billion in 2028. That figure sends an important signal: even as AI stocks face short-term pressure, underlying demand for computing power may remain strong.
The reasons are easy to understand. AI models are becoming more complex, cloud computing services continue to expand, and businesses still need computing resources. Project orders from major cloud providers are providing significant support for data center investment.
But what the market is really worried about is the time lag between investment and returns.
Data center construction requires substantial capital, and power supply, equipment delivery, and regulatory approvals can all affect project timelines. Companies investing tens of billions of dollars today won’t necessarily bring in an equivalent amount of revenue tomorrow.
If AI commercialization doesn’t move fast enough, some developers could find themselves in the awkward position of high investment and low cash flow.
So this AI downturn shouldn’t be simplistically viewed as a golden buying opportunity, nor should it be declared the bursting of a bubble.
What investors should really examine is order quality, debt levels, financing capacity, and future cash flow.
After all, AI can change the world, but what shareholders care about most is: When will it start changing financial results?$MU

‌#美股AI板块遭遇重挫