Chris’ note: As we’ve been looking at this week, the rise of artificial intelligence (“AI”) is impossible to ignore. And it’s about a lot more than just ChatGPT and other AI chatbots.
These “thinking machines” are also helping art experts spot forgeries… detecting cancers on scans more effectively than human doctors… and helping us cure diseases and develop vaccines.
Over at our tech investing e-letter, The Bleeding Edge, colleague Colin Tedards has been showing readers on how to profit. And below, he shares one of his top recommendations for playing the multi-decade boom we see ahead.
And unlike some AI plays that have shot to the moon in 2023, it’s still trading at a reasonable valuation.
We’re at the dawn of a new era…
Since the launch of ChatGPT last November, we’ve seen that computers now not only follow instructions… but also learn.
This means a boom in productivity across the economy as AIs augment the work we do.
It also means gushing profits for companies with AI products and services.
I believe it will be even more lucrative over the next decade than internet stocks such as Amazon, Google, and Facebook (now Meta) were since they came on the scene.
And that’s saying something…
Since they went public, Amazon shares are up 155,600%… Google is up 4,600%… and Meta is up 711%.
Had you invested $1,000 in each at their IPO you’d have $1,610,770 today.
And you didn’t have to buy startups to profit. Microsoft was already a $22 billion company when the first commercial internet browser was launched in 1994.
Had you bought then, you’d be up 12,900%.
Apple was also already a large company when the internet started. It’s up 69,590% over that same time.
That’s why today, I’m sharing one of my favorite AI plays with you. It’s an advanced semiconductor stock that’s primed to profit from the rollout of AI systems at scale.
Sci-Fi Tech Come to Life
ChatGPT can compose poems, pass exams, write movie scripts, and edit documents based on text prompts. It can even write computer code.
It’s sci-fi technology come to life, in other words.
But AI isn’t new…
In 1949, IBM coder Arthur Samuel built an AI system that could play checkers against a human opponent. Since then, we’ve used AI to transcribe speech, guide Mars rovers, assist highway driving, and spot credit card fraud.
And as we looked at here in the Cut this week, AIs are also helping us unlock the secrets of biology… and even spot fake artworks.
You’ve likely been using AI in your own life for years. Personal digital assistants – such as Amazon’s Alexa and Apple’s Siri − are an early form of this technology.
When we ask these AI systems a question, they use a technology known as natural language processing to understand what we’re saying and give us an appropriate answer.
But these AIs all had one specific use case. The developments we’re seeing in AI today are different. For the first time, we can train AI systems on vast datasets for a variety of roles.
This flexibility means that AI is accessible to nearly every business.
Slashing Costs and Boosting Margins
For instance, a major solar panel installer uses AI on satellite images to automate planning. That shaves about 25% off installation costs. So, it can pass some of that savings onto its customers and boost its margins.
Healthcare giant Procter & Gamble is using AI to cut its time and costs to develop new products. The company can ask its in-house AI to create a new soap formula. This saves costs without sacrificing its cleaning factor.
AIs can also help businesses slash costs… and boost their margins.
According to the CEO of IBM, Arvind Krishna, AIs will replace 30% of office jobs over the next five years.
That’s a massive incentive for companies to adopt this technology.
Market research firm Grand View Research reckons the AI software market will grow at 37% annually until it reaches $1.7 trillion by 2030.
And consulting firm PwC says AI will add $15 trillion to the world’s GDP by 2030. That’s a 16% increase over last year’s GDP of $95 trillion.
The question for us as investors: How do we best capture a slice of that mammoth growth?
Innovators and Adopters
When I’m evaluating how companies will benefit from AI, I split them into two groups – innovators or adopters.
The easiest way to get your head around this is to look at the rollout of cloud computing.
Amazon is the perfect example of an innovator in the cloud computing space. It created its Amazon Web Services (“AWS”) division. And it became the first major player to offer cloud computing services.
Until 2017, AWS accounted for more than half the cloud computing market.
Netflix is an example of an adopter. It uses AWS to stream movies and TV shows to its customers.
Rather than create its own cloud computing division, Netflix adopted AWS to cut down on development time. And management quickly realized that AWS could be used to support its streaming ambitions. Netflix was able to scale from 21 million customers in 2011 to 230 million in 2022 with the help of AWS.
The important thing to note is that both categories − innovators and adopters − were hugely profitable.
If you’d bought Amazon stock when it set up its AWS division, you’d be up 7,420% on that investment. And if you’d bought Netflix when it adopted AWS in 2008 for streaming, you’d be sitting on an 11,250% return.
Top AI Innovator
That brings me to today’s recommendation – top AI innovator Advanced Micro Devices (AMD).
It makes semiconductors – or chips – that power ChatGPT and other AI systems.
Rival chipmaker Nvidia is the world’s No. 1 chipmaker for AI. But its shares have already rocketed 212% higher this year. That’s left it priced for perfection.
Nvidia is the undisputed leader in AI with its H100 chips. Others are racing to develop competing chips. But for now, Nvidia controls the market. That means it gets to pick and choose who gets its chips and when.
But here’s what few realize: Nvidia is facing stiff competition. And anybody buying at these levels will almost certainly regret it.
That’s why I like No. 2 AI chipmaker Advanced Micro Devices (AMD). For an opportunity as enormous as this, second place isn’t a bad position to be in.
And AMD is a bargain compared with its better-known rival.
Nvidia trades on a price-to-sales (P/S) ratio of 43. So, investors are pricing the stock at about 43 years of current sales. Maybe Nvidia can grow into that valuation. But it’s about six times the sector average. At that lofty valuation, a misstep could send the stock tumbling.
AMD trades on a P/S ratio of 8.5. So, investors are pricing the stock at a much more reasonable eight and a half years of current sales.
And remember, AMD doesn’t have to dethrone Nvidia to have a great AI business. There’s so much demand for these chips, there’s plenty of room for a close competitor.
And AMD is coming out with a cutting-edge AI chip later this year.
These Chips Make AIs Run Faster
The MI300 chip is an innovation in chip design, architecture, and capabilities. So, it helps AIs run faster and do more tasks.
The MI300 also cuts the space and energy needed to run an AI system like ChatGPT. That saves on the cost to start and maintain an AI data center.
It takes several years of research and development to bring out a new AI chip. So, I expect Nvidia and AMD will have this market largely to themselves through 2024.
Analysts expect AMD to ship 70,000 of its new AI chips this year. About 40,000 of these will go in the El Capitan supercomputer from Hewlett-Packard. It will be the world’s most powerful computer when it goes live.
By comparison, Nvidia is likely to ship more than 400,000 of its H100 AI chips in a single quarter.
But I suspect Amazon will show up as a buyer of MI300 chips for its data centers. Meta could also be a buyer. That would give AMD’s sales a bump.
AMD makes $5 billion to $7 billion in quarterly revenues across all products. If Amazon or Meta become customers for its MI300 chip, this will send AMD sales… and its share price… soaring.
Nvidia may be the king of AI chips. But there’s still plenty of money to be made if AMD can cement its position in second place.
Regards,
Colin Tedards
Editor, The Bleeding Edge