The Chipmaker Behind AI’s Next Big Leap (It's Not Nvidia)

VIEW IN BROWSER

BY KEITH KAPLAN
CEO, TRADESMITH

$200 billion…

That’s how much Microsoft (MSFT) and Amazon (AMZN) are on track to spend on AI this year.

That’s more than the entire annual economic output of Hungary, a country of 10 million people.

The money will fund data centers, AI chips, and the power and networking infrastructure needed to run it all.

That’s great news for the U.S. These companies are building the physical backbone of the AI economy. But it hasn’t been great news for their shareholders.

Since Jan. 1, Microsoft is down 17%. Amazon is down 10%.

The worry is simple. For all their spending, neither company can clearly show investors what they’re getting back. The bills are enormous. But the revenue tied directly to AI remains murky.

That doesn’t mean the AI boom is over – far from it. But it does tell us something has shifted.

In the first phase of the boom, you could make money in obvious names. Microsoft and Amazon for their cloud infrastructure. And Nvidia, the world’s dominant AI chipmaker.

That trade made a lot of people a lot of money. Since ChatGPT launched in November 2022, Nvidia alone has gained 1,011%.

But if this year’s performance is anything to go by, that first phase is coming to an end. The next phase will benefit companies most investors have never heard of, let alone associate with AI.

So today, let’s look at the pattern driving these different phases – and the company best positioned to benefit from what’s coming next.

Recommended Link

I taught my daughter to double the market

That’s right: my Gen-Z, art-student daughter, ran laps around the S&P 500. The year after, she did it again. So, what was the powerful strategy I taught her that allowed her to do better than most money managers on Wall Street? It’s surprisingly straightforward, and incredibly effective. It’s the strategy that earned me the reputation of an “icon” and a “wunderkind” among top investors… It’s the same strategy I use to manage the billions of dollars my clients ask me to manage… And it’s the only strategy I knew I could trust with my daughter’s inheritance. If it’s good enough for my own family, it’s good enough for you. I reveal all the details in this new video, which I urge you to watch today.

The Key to Understanding the AI Boom

Every time a new wave of AI demand hits, it slams into infrastructure that wasn’t built for it. Supply chains scramble. Bottlenecks form. And the companies that fill those gaps hand investors outsized gains.

When AI data centers started multiplying, they needed cooling systems that could handle the heat generated by thousands of processors running flat out.

Vertiv (VRT), which makes thermal management equipment for data centers, was barely on anyone’s radar. Then the orders flooded in – and the stock ran up more than 1,050% in under three years.

Comfort Systems (FIX), which installs HVAC systems for large facilities, climbed more than 1,000% over roughly the same stretch.

Power was the next bottleneck. AI data centers consume electricity at a scale that strains the grid. This led to Bloom Energy (BE), which makes fuel cells for on-site power generation, surging more than 1,230% in under two years.

Then came fiber optics – the hair-thin glass cables that carry data inside and between data centers at the speed of light. The companies that make the lasers, chips, and crystals those cables depend on were suddenly in very short supply. Coherent (COHR) gained nearly 295%. Lumentum (LITE) ran more than 1,160%. AXT (AXTI) climbed more than 1,035%.

Memory was another gap. The surge in AI workloads devoured storage capacity. SanDisk (SNDK) climbed 1,567% in a single year. Western Digital (WDC) gained 579% over two years.

None of these were obvious AI plays when they started moving. None were headline news. But they all filled a new demand gap created by rapid advances in AI – and investors who spotted the pattern early were rewarded.

That pattern is still running. And the emerging bottleneck is Edge AI.

Beyond the Cloud

When you ask ChatGPT a question, your request travels hundreds of miles to a data center the size of several Walmart stores, packed with servers running thousands of specialized AI processors.

The models on those servers process your request and send back an answer.

That’s cloud AI. It works. But it has a fundamental limitation. It needs a high-speed internet connection.

And for certain applications – a car navigating traffic, a factory robot making split-second decisions, a medical device monitoring a patient’s heart – waiting for a round trip to a distant server isn’t just inconvenient. It’s unacceptable.

That’s where Edge AI comes in. Instead of sending data to the cloud, these systems process information right where it’s collected – inside the device itself.

A self-driving car makes hundreds of decisions a second. It can’t wait for a server a thousand miles away to weigh in. Neither can a factory robot. Or a hospital monitor.

And like every previous development in this boom, Edge AI is creating a new demand gap.

The chips that power cloud AI – Nvidia’s data center GPUs – aren’t designed for this. Edge AI needs something different. Smaller. More power-efficient. Rugged enough to survive a car engine or a factory floor. Built for real-time decisions, not bulk data processing.

One company above all others fills that gap.

A Chipmaker for AI’s Next Frontier

NXP Semiconductors (NXPI) doesn’t make headlines the way Nvidia does.

If you looked it up on a financial site, you’d find it described as a semiconductor company focused on automotive and industrial applications.

That description is accurate. But it misses something important.

NXP makes the chips that go inside vehicles and factories – precisely the environments where Edge AI is being deployed right now. Its processors handle the real-time decisions that keep a car in its lane, coordinate a robotic assembly line, and manage industrial sensors without ever pinging a cloud server.

Automotive and industrial are the two largest markets for Edge AI. They’re also the two markets where NXP has spent decades building relationships.

This may be a new type of AI play. But it isn’t a moonshot. At roughly $62 billion in market cap, NXP has the financial stability to weather the volatility that comes with being in the middle of a technology transition.

Trillions Will Flow Into These Hidden AI Stocks

NXP isn’t the only AI company flying under the radar. It’s part of a group of “hidden” AI plays that are set to become the next trillion-dollar stocks.

On the surface, they don’t look like AI stocks. But they’re positioned to profit from the trillions of dollars that will hit different areas of the AI industry.

That’s why Jason Bodner – our money flows expert here at TradeSmith – is teaming up with former tech executive, angel investor, and early AI backer Jeff Brown for a special event all about how to find these hidden AI plays.

Jeff recommended Nvidia in 2016… AMD in 2017… Tesla in 2018… and he’s personally invested in nearly a dozen unicorns and even decacorns – startups worth $10 billion. He also an executive at NXP Semiconductors.

Now, he’s teaming up with Jason to showcase a system for spotting these stocks before they become household names. They’ll each also unveil a “secret” AI stock you can buy to profit from the next phases of the AI boom.

They’re getting into all the details next Wednesday, Feb. 25, at 8 p.m. ET. You can go here to add your name to the guest list.

The Chipmaker Behind AI’s Next Big Leap (It's Not Nvidia)