News Room


Nvidia is already the third-most-valuable company on the planet, trailing only Microsoft and Apple. It’s the undisputed leader in the production of chips essential for powering the generative artificial intelligence revolution.

Companies are throwing as much money as possible at Nvidia in order to stockpile chips in datacenters that will churn out unique images, videos, text and other content in the years ahead.

Shares of the company have surged by 65% year-to-date, just two months into the year. The gain in Nvidia shares alone this year has powered 32% of the gain in the S&P 500.

It’s a stunning achievement but for all that, shares aren’t that expensive. They’re trading at 32x forecast earnings in the next 12 months and have grown earnings at a pace never seen before at any company (in dollar terms at least). For context, companies during the dot-com era were trading at 60-100x earnings; so on that alone, shares could more than double.

What could really propel its growth further?

NVDA daily

Looking at the chart, it’s incredible (particularly as a guy who owned shares in 2011, but hasn’t since then unfortunately). It’s tempting to think that it will all come crashing down. The company is running with +75% margins right now and they don’t even manufacture their own chips. Surely someone will catch up by the end of the decade with something at least comperable to what they’re offering, right? The long history of chips shows that it’s more of a commoditized product than a moat.

Maybe. But let’s set that all aside for a minute.

Bubbles happen because of mass psychology. A mania mindset emerges as people come to grips with a world-changing idea. All the pieces are already in place for that with Nvidia and its rise from $150 at the dawn of ChatGPT to $818 now captures much of that.

But is 5x really a bubble in the company that’s clearly the leader in AI? Especially when actual earnings have matched the growth rate of the stock?

I would argue not. This could be just the beginning. Mass psychology is hard to predict but here’s the line of thinking that could lead Nvidia much higher:

Virtually all invention in the future will be via AI

Human invention is a relic. Already we’ve seen phamaceuticals that were developed via AI and it’s just the tip of the iceberg. With good data (and that might not be easy to find), generative AI can unlock the molecular mysteries of the human body and how we can enhance and heal ourselves in ways never before considered. AI is going to cure cancer and so much more.

That’s just in one field. It could do the same in materials, design, engineering, accounting, programming and many more. Within that, whoever has the largest army of chips owns those inventions. It’s the key to unlocking the future and it will be winner-take-all.

Moreover, the stakes may even be higher for governments as generative AI is tasked with weapons design, including biological, chemical and nuclear weapons along with defenses against those things. What’s that worth to the US, China, Russia, Iran and North Korea? It will become a national security priority to amass the largest bank of AI chips possible, with no cost being too high.

Finally, it will be AI designing the next generation of computer chips. Nvidia’s main task will be using its own chips to create the next generation of chips and so on. It will also use that power to design custom chips for clients, something it’s already working on.

“The chip industry is the foundation of nearly every other industry in the world,” said Jensen Huang, founder and CEO of Nvidia.

Now do I believe all of that will come to pass? I can certainly see some holes in that argument. Do I think that enough people will believe in that idea to create perhaps the biggest single-stock bubble in history? Enough to make Nvidia the most-valuable company in the world (it would only need to rise 50% from here)?

I think that very soon people will be screaming that same argument as Nvidia crosses $1000/share and beyond.


Source link

Leave a Reply

Your email address will not be published. Required fields are marked *