Here Are 5 AI-Related Stocks to Buy and Hold for the Next 5 Years

The major driver of the stock market right now, and a huge contributor to economic growth, is the massive investment in artificial intelligence (AI) infrastructure -- primarily AI data centers -- by the so-called hyperscalers.

Hyperscalers are enormous technology companies that are borrowing and spending hundreds of billions of dollars a year to build AI data centers, including the big four: Meta Platforms, Microsoft, Amazon, and Alphabet. There are others, of course, but those four have more than tripled their capital expenditures (capex) on AI infrastructure over the last five years.

Companies worldwide spent almost $1 trillion on data centers last year, according to McKinsey & Company. That's projected to reach $4 trillion by 2030.

I would be surprised if that figure is not revised higher over the coming months and years -- many times. As Apollo noted in a recent research report, estimates of capital expenditures on data centers have risen multiple times over the past year.

Image source: Getty Images.

If those forecasts are even partly true, AI capex will continue to drive economic growth and sustain the bull market for several years.

I wouldn't attempt to predict which of the four companies above will win this arms race to be the dominant provider of AI compute. But I am confident that the construction, maintenance, and ongoing scaling of these AI facilities will require substantial inputs. Namely, electricity, memory capacity, copper, and certain types of electronic components.

So, I've come up with a short list of companies providing those inputs, most of which are already seeing their revenue, profits, and share prices soar due to demand from data center spending. These should all be great stocks to buy and hold for at least the next five years.

Copper is indispensable, and demand is soaring

First, there's copper. A traditional data center requires between 5,000 and 15,000 tons of copper. AI data centers, in contrast, can need up to 50,000 tons of copper per facility, according to the Copper Development Association. As a result, the spot price of the red metal has increased from about $486 per pound one year ago to around $656 a pound today. That's a 35% increase.

A great way to play that is the Global X Copper Miners ETF (NYSEMKT: COPX), which tracks an index of global companies engaged in copper exploration, mining, and refining. It has more than doubled in price over the past year.

Next is power. Data centers are a huge new burden on the electric grid and are driving significant investment to bring more power online. The World Resources Institute estimates that global transmission and distribution infrastructure will need to double by 2050 to accommodate the world's power needs. GE Vernona (NYSE: GEV) is a leader in gas and steam turbines for power plants, wind turbines, and grid components. The stock is up 85% over the past year.

AI data center demand for memory is insatiable

The third major input is memory. Data centers have an insatiable need for it. Micron Technology (NASDAQ: MU) makes the dynamic random-access memory (DRAM) and NAND memory they need, and there's a global supply shortage, which has sent prices of DRAM and NAND chips soaring. As a result, the stock is up more than 900% over the past 52 weeks and recently entered the $1 trillion market-cap club.

Sandisk (NASDAQ: SNDK) makes the flash memory products that data centers also need in large volumes. Its products are generally solid-state drives that "hold" the memory so other chips, such as DRAM, can access it. The stock has soared more than 650% so far in 2026.

Certain other electronic components are also necessary for any data center. Taiyo Yuden (OTC: TYOYY), a Japanese company, manufacturers multilayer ceramic capacitors (MLCCs), which regulate power flow in electronic devices and are a critical component in data centers. A single Nvidia circuit board can require more than 6,000 MLCCs, and there is currently a global supply shortage of them. Goldman Sachs expects MLCC demand from AI servers to at least quadruple by 2030, while industry capacity is rising by about 10% per year. Taiyo Yuden's share price has soared more than 550% over the past year.

Finally, a bonus sixth pick for this AI-infrastructure inputs portfolio, and this one is rarely thought of as an AI-related company. I'm talking about Caterpillar (NYSE: CAT), maker of construction and farm equipment. Each data center construction project requires significant excavation, land clearing, and construction equipment, and CAT has the size and scale to handle the new level of demand. The stock is up 165% over the past year.

There you have it. A six-stock AI pick-and-shovel portfolio for the next five years.

Matthew Benjamin has positions in Alphabet and Microsoft. The Motley Fool has positions in and recommends Alphabet, Amazon, Caterpillar, GE Vernova, Goldman Sachs Group, Meta Platforms, Micron Technology, Microsoft, and Nvidia. The Motley Fool has a disclosure policy.