The Canary In The Coal Mine: Is Nvidia The First Sign Of The AI Bubble Bursting?
The Canary In The Coal Mine: Is Nvidia The First Sign Of The AI Bubble Bursting?
In short, yes.
Over the past few years, there has been an ever-growing chorus of people pointing out the inane price of AI stocks and claiming a market bubble is ready to burst and possibly even take out some big tech giants. AI diehards have mostly been able to brush off these worries, claiming the technology is poised to revolutionise practically every industry on Earth, justifying said stock prices. Well, we may have had our first tangible signs that there is not only an AI bubble but also that it is starting to burst. Nvidia recently announced it smashed its earnings predictions and hit a record revenue of $30 billion over three months. In fact, Nvidia doubled its sales compared to the same period last year. This means Nvidia now controls over 80% of the AI chip market, solidifying the company as a monolith of the AI industry. Yet, despite all of this, Nvidia stock fell 6% after this announcement, and AI companies associated with Nvidia, such as Meta and Amazon, also took a hit. The reasons for this counterintuitive slip paint a textbook example of a market bubble and how they burst.
But before we dive into that, let’s actually define a market bubble.
A market bubble happens when investors pour money into a stock or stock associated with an industry, not based on solid business fundamentals, inflating its value way past where it should be. Great examples of this are the tulip mania of the 1600s, Japan’s real estate bubble in the 1990s, and the dot com bubble of the early noughties.
But bubbles don’t last and eventually burst. This happens when some market correction awakens investors to the reality of the businesses/industry they have invested in, leading to a mass stock sell-off, causing a crash that can impact other linked industries or even entire economies. Bursts aren’t always sudden and can happen over the course of a year or two, particularly if giant corporations have invested. But either way, if there is a bubble, a burst is inevitable.
So, is the AI industry in a bubble? Well, Nvidia is the largest AI company at the moment, and its stock price has grown by 3,000% since 2019, but its revenue has only grown by 60% since 2019. It does have new technology coming; however, it is by no means a giant leap from its current offerings. As such, Nvidia’s stock price is highly indicative of a bubble. This is true for many other AI-based stocks, such as Tesla.
So, what spooked the investors? Surely, record-breaking revenue should have emboldened them to continue inflating this stock. Well, we don’t have to guess, as market analysts have already told us. Simon French, head of research at Panmure Liberum, said that Nvidia’s sales growth was simply not enough. He claims investors have seen Nvidia smash revenue targets year after year, and this time, Nvidia only broke its target by a small amount. He told the BBC, “If you’re going to raise expectations that high, then you’ve got to keep growing at spectacular rates.” Matt Britzman, senior equity analyst at Hargreaves Lansdown, echoed this notion, saying, “It’s less about just beating estimates now. Markets expect them to be shattered, and it’s the scale of the beat today that looks to have disappointed a touch.”
In short, these investors sold off Nvidia stocks as it didn’t meet their expectations, despite doubling its revenue compared to the same period last year! Such lofty expectations and exiting so quickly are typical of a bubble.
But it might be more than just the lower-than-expected revenue that caused this sell-off. You see, Nvidia’s business model of selling AI chips has been brought into question.
Why? Well, firstly, OpenAI, one of Nvidia’s biggest customers, has warned that next-generation AI models will use so much energy to develop, train and maintain that they will require an energy breakthrough like nuclear fusion to be viable. As such, they will likely not continue to buy exponentially more chips from Nvidia as they have been. That is if OpenAI is still around. The AI pioneer is set to post a loss of $5 billion by the end of the year and could face bankruptcy! If Nvidia lost such a major customer, it could significantly impact its revenue and stock price.
But, OpenAI has the largest revenue of any AI service provider. So, if it can’t make a profit, who can?
Sequoia Capital Partner David Cahn also asked this question. Over 60% of Sequoia’s recent investments have been in AI companies, so they have a huge interest in these companies succeeding. However, Cahn found that by the end of this year, AI companies, such as Google, Meta, Microsoft, AWS, and OpenAI, are set to spend $150 billion on AI chips, mostly from Nvidia. However, Cahn found that running these chips will cost around $150 billion a year. Cahn also found that profit margins need to hover around 50%, as this generates enough funds to pay for AI development, new infrastructure, hiring better talent, and so on. As such, to see a return on their massive chip expenditure, the major AI companies must generate an additional $600 billion of revenue or more than three times the industry’s current total revenue.
Now, remember that AI development has hit diminishing returns, causing stagnation despite ever-increasing R&D expenditure and the fact that AI companies have admitted that next-gen AI models are currently unviable. With all of this in mind, do you really think that Nvidia can keep up its record-breaking chip sales? I certainly don’t.
I suspect a good number of investors, many institutional investors, having noticed that Nvidia’s sales are already starting to not grow as fast as they thought, think the bubble will burst soon and are cashing out before this house of cards falls. This is most likely the first sign of the bubble bursting. I can’t see how it can be anything else.
Thanks for reading! Content like this doesn’t happen without your support. So, if you want to see more like this, don’t forget to Subscribe and help get the word out by hitting the share button below.
Sources: Will Lockett, Will Lockett, BBC, Sky, The Independent, Statista