Goldman Sachs Says a $7.6 Trillion AI Spending Boom Is Coming. Skip the GPUs, Follow the Money Here
Goldman Sachs sees hyperscalers accelerating toward a spending level that dwarfs anything the tech industry has attempted before, and the companies positioned to profit most may have nothing to do with the chips getting all the attention.
This post may contain links from our sponsors and affiliates, and Flywheel Publishing may receive compensation for actions taken through them.
Artificial intelligence has moved beyond a technology spending cycle and into an infrastructure buildout measured in trillions of dollars. Goldman Sachs estimates that the five largest hyperscalers will spend about $800 billion on capital expenditures in 2026, up 94% from 2025. More importantly, that spending isn’t expected to fall. The growth rate is simply coming down from an extraordinary surge.
Goldman projects hyperscaler capex will reach $1.2 trillion in 2027 and $1.4 trillion in 2028. That changes the investment opportunity. The first AI boom rewarded companies selling the processors. The next stage could reward the businesses supplying everything required to keep those processors running.
The AI Spending Boom Is Getting Bigger
To put those numbers in perspective, the five largest hyperscalers went from roughly $150 billion of annual spending in 2023 to an estimated $800 billion in 2026. By 2028, Goldman expects that figure to reach $1.4 trillion.
The additional $200 billion expected in 2028 alone is almost as much as these companies spent during all of 2024. And the $1.2 trillion projected for 2027 exceeds their combined spending from 2023 through 2025.
This isn’t an AI infrastructure cycle running out of gas. It’s beginning to look more like a city being built around a gold mine — and the next fortunes may belong to the companies supplying the roads, electricity, and buildings.
The first AI boom rewarded companies selling the processors. Now that has become the easy part. The harder effort is supplying the electricity, networking, cooling, memory, and physical infrastructure required to turn those chips into productive computing capacity.
Goldman Sachs’ Global Institute and Global Investment Research estimate roughly $7.6 trillion will be invested in AI infrastructure from 2026 through 2031, covering compute, data centers, and power. Its baseline model reaches $1.6 trillion of annual AI capex by 2031.
The Bottleneck Is Moving Beyond GPUs
The obvious AI winners have been chip companies. But a GPU sitting in a warehouse produces exactly zero AI revenue.
It needs networking, memory, optical connections, cooling, electrical equipment, buildings, and enormous amounts of electricity. President Trump called it “the oil of the next 50 years.”
The International Energy Agency expects global data-center electricity consumption to roughly double from 485 terawatt-hours in 2025 to 950 TWh in 2030. Electricity consumption at AI-focused data centers is projected to triple.
That creates a different roadmap for investors. The first AI spending wave was dominated by computing power. The next one is likely to be shaped by three bottlenecks that become harder to ignore as hyperscalers deploy hundreds of billions of dollars more each year.
| AI Bottleneck | Sector Leaders | Why They Matter |
| Memory | Micron (NASDAQ:MU | MU Price Prediction), SK hynix (NASDAQ:SKHY), Sandisk (NASDAQ:SNDK) | AI accelerators require enormous amounts of high-bandwidth memory, while expanding data centers also need storage. |
| Networking & Optics | Broadcom (NASDAQ:AVGO), Arista Networks (NYSE:ANET), Coherent (NYSE:COHR) | As AI clusters grow, moving data between thousands of processors becomes as important as processing it. Networking chips, switches, optical transceivers, and lasers provide the connections. |
| Power & Electrical Infrastructure | Eaton (NYSE:ETN), GE Vernova (NYSE:GEV), Vertiv (NYSE:VRT) | Every new AI cluster needs electricity, power distribution, transformers, cooling, and other equipment before a single GPU can do useful work. |
There are other ways to play the buildout, including energy storage, power generation, and data-center construction, but investors don’t need to look far beyond these three bottlenecks to find where the next wave of AI infrastructure spending is likely to flow.
The IEA also says supply chains for gas turbines, transformers, advanced chips, and IT components have tightened as data-center projects expand.
That’s the key point: AI increasingly needs the physical economy.
Follow the Money Into the Physical Economy
Goldman Sachs says strong demand and constrained supply have already pushed memory producers’ gross margins to roughly 80%, more than twice their historical average. That also illustrates the risk: today’s bottleneck profits can attract tomorrow’s capacity.
Granted, investors shouldn’t assume every AI infrastructure stock will prosper. Goldman notes that slower capex growth will eventually meet rising depreciation expenses, potentially reducing AI’s contribution to S&P 500 earnings.
Key Takeaway
In short, the AI trade is getting wider, not ending. The numbers point toward $1.4 trillion of annual hyperscaler spending by 2028 and $7.6 trillion of cumulative AI infrastructure investment through 2031.
Smart investors don’t have to chase every GPU cycle. The more durable opportunity may be owning companies that sell the electricity, cooling, networking, memory, and electrical equipment that $1.4 trillion of annual spending cannot avoid.
Contact [email protected] for any questions or corrections.
After two decades of patrolling the dark corners of suburbia as a police officer, Rich Duprey hung up his badge and gun to begin writing full time about stocks and investing. For the past 20 years, he’s been cruising the markets looking for companies to lock up as long-term holdings in a portfolio while writing extensively on the broad sectors of consumer goods, technology, and industrials. Because his experience isn’t from the typical financial analyst track, Rich is able to break down complex topics into understandable and useful action points for the average investor. His writings have appeared on The Motley Fool, InvestorPlace, Yahoo! Finance, Money Morning, and, of course, 24/7 Wall St. He has been featured in both U.S. and international publications, including MarketWatch, Financial Times, Forbes, Fast Company, and USA Today.
