Can you guess the top-performing stock on the S&P 500 so far this year?

If you guessed Nvidia (NVDA), you are – as of Wednesday – incorrect. That title now belongs to Vistra (VST), up some 200% on the year as I write this.

What is Vistra? What does it do? And how is the stock beating almighty Nvidia?

All of that in due time. But first, have a look at this.

Bloomberg (emphasis added):

OpenAI has pitched the Biden administration on the need for massive data centers that could each use as much power as entire cities, framing the unprecedented expansion as necessary to develop more advanced artificial intelligence models and compete with China.

Following a recent meeting at the White House, which was attended by OpenAI Chief Executive Officer Sam Altman and other tech leaders, the startup shared a document with government officials outlining the economic and national security benefits of building 5-gigawatt data centers in various US states, based on an analysis the company engaged with outside experts on.

For some context, a nuclear powerplant producing 1 gigawatt per hour (gigawatt-hour) would be able to power 876,000 American homes per year. As Bloomberg helpfully points out, 5 gigawatts are roughly equivalent to five nuclear reactors.

Speaking of nuclear power plants, that brings us back to Vistra, which happens to be an electricity and power generation company with a strong portfolio of nuclear power plants.

What does all this mean?

Let’s have a look…

$1 Trillion in Capex

When looking at the price tag associated with the development of artificial intelligence, a recent paper from Goldman Sachs provides an eye-popping estimate.

From the report:

The promise of generative AI technology to transform companies, industries, and societies continues to be touted, leading tech giants, other companies, and utilities to spend an estimated ~$1tn on capex in coming years, including significant investments in data centers, chips, other AI infrastructure, and the power grid.

I’m always skeptical when I see nice round numbers like that. But as a ballpark, that’s probably not far off. It will likely end up being higher. And the capex numbers coming out of the big techs would seem to prove it.

From their most recent earnings, the large tech companies have reported capital expenditures of…

  • Microsoft: $13.8 billion

  • Alphabet: $13.1 billion

  • Meta: $8.17 billion

And, yes, most of this capital is going towards AI and data center-related projects, as Microsoft admitted on its latest earnings call:

Cloud and AI related spend represents nearly all of total capital expenditures. Within that, roughly half is for infrastructure needs where we continue to build and lease datacenters that will support monetization over the next 15 years and beyond.

And they’re not slowing down anytime soon:

To meet the growing demand signal for our AI and cloud products, we will scale our infrastructure investments with FY25 capital expenditures expected to be higher than FY24.

And as Brad has been sharing with you for months, all this means one thing: More energy demand… lots more.

So much so that Microsoft recently signed a 20-year contract with Constellation Energy (CEG), the owner of the infamous Three Mile Island nuclear plant.

As Brad shared earlier this week:

Constellation will spend approximately $1.6 billion to get the plant back on its figurative feet. And Microsoft will purchase the resulting energy for its artificial intelligence (AI) power generation between 2028 and 2048.

And if some of the more ambitious AI plans come to fruition, energy demand will only increase. For instance, it’s long been rumored that OpenAI CEO Sam Altman has been angling towards a mega-project called “Stargate.” The Stargate project would combine millions of XPUs (multiple data centers) to create a veritable supercomputer.

The goal of the project is to achieve the holy grail of AI: artificial general intelligence (AGI). And analysis I’ve seen points to figures around $57 billion over the next five years in hardware costs alone to get the project off the ground.

That’s a big swing. But with $88 billion in net income over the past 12 months, Microsoft is swinging a big bat. If it wanted to, it could probably get it done.

And ask yourself this: Will Alphabet, Meta, Apple, or Tesla let Microsoft run away in the AI race?

Not a chance…

If these numbers are even close to accurate, this is more than just another investing trend. We’re staring down the barrel of a multi-year flood of spending and building.

And so, the question worth asking is: What’s an investor to do about it?

We Still Like Energy

In June, Brad published a recommendation for WEC Energy (WEC) in the pages of The Wide Moat Letter. As he said at the time:

Andrew Carnegie understood that – no matter what happened to individual companies – the trend of global industrialization relied on one essential ingredient, steel.

Right now, many investors are trying to guess which firm will dominate the AI trend. For instance, can Nvidia really maintain its dominant position – with 75% gross margins – for years to come?

It seems unlikely. The company has an incredible head start. But semiconductors are a notoriously competitive industry. Sooner or later, firms will find a way to eat away at Nvidia’s market share. There’s simply too much money at stake.

Many investors are trying to guess which AI firm will win out in the years ahead. We’re interested in asking a different question: What is AI’s “steel?”

The answer, of course, is energy. And the recommendation for WEC was timely. That position is recording an 18% return in just over three months. That’s impressive for a humble utilities company.

We still like that thesis. While the large techs battle for AI supremacy, the common denominator for all of them is energy. And, ideally, it would be nuclear energy.

Brad and the entire team are on the lookout for compelling ideas in the nuclear energy space. The only problem we see is that there are few SWAN (sleep well at night) options at the moment.

Vistra has had an incredible run. But we wouldn’t feel comfortable recommending it at current levels. The same goes for a company like Constellation Energy (CEG), up 122%, year-to-date.

But rest assured, if we find another attractive investment tied to energy production for data centers, Wide Moat subscribers will be the first to hear.

Regards,

Nick Ward
Analyst, Wide Moat Research

MAILBAG

What are your thoughts on the action stirring in the nuclear energy space? Write us at [email protected].