Nvidia (NASDAQ:NVDA) is now worth over $1 trillion. But in recent days NVDA stock has been backing up.
Since hitting a peak on June 20, it is down almost 5%. The market cap is still over $1 trillion entering trade June 27, but barely.
Nvidia’s graphics processing units, which focus on math rather than executing program instructions like central processing units, are at the heart of its rise. But Nvidia has also built a suite of software around its hardware, allowing clouds to quickly execute Generative AI tasks.
It’s these tasks, whether outputting pictures, words, music, or computer code, that have become the big tech story of 2023. But NVDA stock is now selling for more than 30 times its revenue, over 200 times its earnings. How much is too much?
A Close Look at NVDA Stock
How did Nvidia leave such a host of competitors in its wake? No semiconductor company has even been this valuable. Advanced Micro Devices (NASDAQ:AMD), for example, is worth $180 billion. Intel (NASDAQ:INTC) is worth $140 billion.
The answer is software.
Nvidia knows its designs must be turned into chips by others. That’s why it built a platform around them. As Moore’s Law advanced toward measuring circuit distances in Angstroms, value has moved from manufacturing to design, from hardware to software.
Nvidia chips and software are now a must-have for any company supporting Generative AI applications.
Clouds, originally built around the cheapest possible processors, are now being upgraded using expensive Nvidia chips. Supply is limited by Taiwan Semiconductor’s (NYSE:TSM), manufacturing capacity, so prices are high.
Thanks to its AI platform, Nvidia has become both the Intel and the Microsoft (NASDAQ:MSFT) of the AI revolution. This is what brings Nvidia its value.
How High the Sky?
I didn’t see the current mania coming, and I have hedged my bets. I own NVDA stock as well as shares in Intel, AMD and TSM. (TSM is now worth $520 billion.)
But it was applications built around the “WinTel” platform that created value in the 1990s. Investors who know their history, like Cathie Wood, are now switching their bets toward these applications.
Wood has been selling Nvidia into strength and buying companies like UiPath (NASDAQ:PATH) Twilio (NASDAQ:TWLO) and Teladoc Health (NASDAQ:TDOC).
The danger in this strategy is that it’s hard to know who a revolution’s winners will be in advance. I’m old enough to remember when Novell ruled networking in the early 1980s. I remember when Salesforce.com (NYSE:CRM) was a gnat next to Oracle (NASDAQ:ORCL).
Nvidia itself is pushing the results obtained through deep learning by small companies like Chooch, and by universities with no product to sell.
But when a balloon is near the top of its arc, any stiff breeze can send it down. The recent weakness in Nvidia stock is being blamed on downgrades of Tesla (NASDAQ:TSLA). Does it make any sense? No. But trading patterns don’t have to make sense.
The Bottom Line
The AI bubble is in its early stages.
Like the PC boom, the multimedia boom, and the Internet boom before it, we don’t know who the ultimate winners will be. That is, other than the companies that started it all.
In the case of AI, that company is Nvidia. Until someone starts making bank on AI applications, NVDA stock will be a safe harbor, as International Business Machines (NYSE:IBM) was through the 20th century.
But as IBM’s fate illustrates, the best investment strategy is to spread your bets. You never know when a company might lose the mandate of heaven. Stay with NVDA stock but keep your eyes open for growth and profit in the applications space.
As of this writing, Dana Blankenhorn had LONG positions in NVDA, MSFT, TSM, INTC, CRM, and AMD. The opinions expressed in this article are those of the writer, subject to the InvestorPlace.com Publishing Guidelines.