The AI Stocks Hiding In Unavoidable Metals Like Copper And Uranium

The story of artificial intelligence and data centres has captured the focus of investors around the world.

As a result, many investors have turned their attention to semiconductor, artificial intelligence and data centre companies.

But what about what those companies require to operate?

We’ve previously discussed the importance of energy and cooling to the operations of data centres. However, both of those have things in common, namely, they require critical minerals and metals to actually transmit the energy and cool the servers.

Two of the most important substances are copper and uranium.

Why AI data centres can’t function without copper

Just as blood keeps the body alive, copper keeps a data centre running, carrying the power that fuels its every function.

It sends power from the grid to the data centre, it can be found in the chips, powerboard, electrical connecters, power distribution and even the cooling. Due to its high electrical connectivity and heat management, it is present in almost every aspect of a data centre’s operations.

For example, in 2008, Microsoft built a US$500 million data centre in Chicago which experts estimate required around 2,177 tonnes of copper, or 27 tonnes per megawatt of applied power. Given how much larger data centres have become since 2008, especially those used for artificial intelligence, some estimates predict hyperscaler data centres could require as many as 50,000 tonnes of copper.

Data centres also need large amounts of power to function. Both the generation of this power and its delivery to the data centre require copper.

Such is the growth in data centres, especially in regard to the expansion of artificial intelligence and machine learning, and the sheer hunger they have for energy, some estimates have data centres consuming as much as 12% of total US energy production by 2028.

As a result, mining giant, BHP, believes that copper used in data centres will grow six-fold by 2050.

Why nuclear energy could be the beating heart of AI

If copper is the blood flow of AI data centres, uranium and nuclear power could be the beating heart.

More and more tech giants are turning to nuclear power to act as the core source of energy to fuel their increasingly larger data centres.

For example, Amazon has invested money into developing small, modular, nuclear reactors to help power its data centres. Google has done the same. Also, in 2024, Constellation Energy, on behalf of Microsoft, reopened one of the reactors at Three Mile Island, a place better known as home to the worst nuclear accident in US history, although that was a different reactor to the one Microsfot is using.

The reason for the interest in nuclear energy, and there for uranium, is simple.

Nuclear is an extremley reliable, low-emission and efficient source of energy.

A nuclear reactor is reliably able to be run 24/7 and rarely needs to be shut down. This is important given that data centres act as the critical infrastructure that allows digital traffic to flow. If a data centre shuts down, it also essentially shuts down all the websites, cloud services and AI functionality that is built on top of it.

Nuclear also has the highest capacity factor of all energy sources, with the US Department of Energy estimating its capacity factor, or how much a plant is able to be run at maximum power, at around 92%. For example, coal power plants only rated around 43% with natural gas at almost 60%.

With low-emissions and the ability to offer consistent, reliable, energy, it is not a surprise that it is starting to be seen as the power source of choice for AI data centres, which also means according to many experts, an increased demand for uranium to fuel those nuclear power plants. 

Which copper and uranium stocks are helping to power the AI revolution?

Copper and uranium are yet more examples, just like our previous articles about energy and cooling, of potential investment opportunities which don’t immediately scream 'AI stocks' at first glance.

However, there are institutions which are paying close attention to the theme.

Take Global X’s Artificial Intelligence Infrastructure ETF (ASX:AINF) for example.

Rather than invest in the typical tech names, it has instead built a portfolio around the types of companies which help create the underlying infrastructure that allows artificial intelligence to function. Copper and uranium make up a large part of this portfolio.

In fact, its second-largest position is the US copper company, Southern Copper (SCCO). The portfolio also includes copper miners like Freeport- McMoRan (FCX), and Teck Resources (TSX:TECK.B) amongst others. It also includes uranium companies like Cameco Corp (TSX:CCO) as well as Constellation Energy (CEG) who is the company mentioned earlier that reopened a reactor in Three Mile Island for Microsoft.

Between these companies, it is easy to see why some consider this portfolio a good selection of companies who could solve AI’s power problem.

What does it mean for investors?

Artificial intelligence is more than just a software platform, or even code, that allows for problems to be solved.

For AI to work, it requires the reliable operation of infrastructure spanning multiple industries to be working in sync, whether this is chips, servers, cooling and, yes, even power generation and the supply of critical minerals.

If any of these components fail, or even not work as effectively as desired, than the impact of artificial intelligence could be significantly impacted.

In the case of copper and uranium, we may have two metals which may have the biggest say in regard to the growth of artificial intelligence and the infrastructure required to fuel that growth and every innovation that may come from it.

Explore why the Global X AI Infrastructure ETF could be 19% undervalued.

Have feedback on this article? Concerned about the content? Get in touch with us directly. Alternatively, email editorial-team@simplywallst.com

Simply Wall St analyst Andrew Legget and Simply Wall St have no position in any of the companies mentioned. This article is general in nature. We provide commentary based on historical data and analyst forecasts only using an unbiased methodology and our articles are not intended to be financial advice. It does not constitute a recommendation to buy or sell any stock and does not take account of your objectives, or your financial situation. We aim to bring you long-term focused analysis driven by fundamental data. Note that our analysis may not factor in the latest price-sensitive company announcements or qualitative material.