The AI Bubble Begins to Burst: What’s Behind the Slump on Wall Street?
Doubts Grow on AI Trade as Wall Street Weighs Risks of an "Arms Race"
A growing number of strategists and hedge-fund founders seem downbeat on the artificial intelligence trade, with concerns about circular financing and debt-fueled spending weighing on markets. Despite recent rallies, doubts about the durability of the AI narrative were having a moment on Wall Street this week.
Several high-profile analysts published reports focused on potential risks to the AI trade, including circular financing in AI and data-center commitments. Recent deals between Oracle Corp., OpenAI, and Nvidia Corp. revived concerns about where money for these new commitments would come from. For the first time, companies appeared willing to take on more debt to get in on the action.
Oracle’s "unusual" jumbo bond deal early this week reflected a new trend – companies taking on large sums of debt to finance their participation in the AI trade. Oracle has broken the pattern followed by other major players like Microsoft, Meta, Alphabet, and Amazon, which could afford to pay out of cash flows.
A Market in Flux
As doubts about the AI trade grow, investors’ eagerness to buy on dips is still limiting losses. Mark Hackett, chief market strategist at Nationwide, noted that while major indexes finished just shy of record territory this week, several AI-related names came under pressure.
The Nasdaq Composite COMP was on track to finish lower for a fourth straight day earlier this week, its longest string of daily declines since April. However, it managed to close the session in the green. "At some point, this has to translate into the bottom line," warned Steve Sosnick, chief strategist at Interactive Brokers.
The AI Spending Boom and the Telecom Bubble
Sosnick drew parallels between the current AI spending boom and the rush to lay fiber-optic cable that fueled the telecom bubble in the late 1990s. That episode ended with high-profile bankruptcies and scandals as companies struggled with excess capacity.
A team of equity analysts at Barclays also explored risks associated with the AI trade, including what if power constraints make it more difficult or impossible to continue building data centers? What if AI models stop improving, leading to excess capacity?
The term "dark fiber moment" was mentioned – a reference to the telecom bubble where companies built excessive infrastructure as expectations soared for internet traffic. This ultimately led to initial underutilization of these assets.
An Arms Race in the Making
J.P. Morgan Asset Management’s Michael Cembalest highlighted potential constraints on the AI build-out, including aggressive demand for power, which has already raised electricity prices for consumers in certain markets.
Since ChatGPT launched in November 2022, a group of 41 stocks associated with artificial intelligence have driven 75% of the S&P 500’s advance. They have accounted for 80% of corporate-earnings growth and 90% of capital spending. However, emerging financial risks are factoring into the discussion.
The Oracle-OpenAI deal broke the trend of AI-related spending being primarily financed by companies like Microsoft, Meta, Alphabet, and Amazon out of their robust cash flows. With this new bond issue, Oracle has potentially ignited an "arms race" among other companies to borrow heavily to participate in the data-center build-out.
Risk Factors
Cembalest also pointed out the risk that debt-fueled spending could supercharge the build-out and leave some investors holding stranded assets if many data centers end up being underutilized. This echoes concerns raised by other experts, including Citadel’s Ken Griffin and Greenlight Capital founder David Einhorn.
An article published this week in the Wall Street Journal questioned whether AI spending would ever pay off, while also warning that unchecked data-center spending could devastate large sums of capital.
Technical Signs
Chart readers can see signs that the AI trade is becoming stretched. The Global X Artificial Intelligence & Technology ETF (AIQ) reached a 14-day relative-strength index (RSI) of 82.41 on Monday – its most overbought level since inception, except for one previous reading in June 2023 when it peaked at 80.80.
Not Just Another Bubble
Doubts around AI aren’t new, and some Wall Street players have flagged these concerns for months or even a year ago. However, Oracle’s decision to tap the debt markets has revived scrutiny this week. Some experts are now cautioning that investors need to consider how this "arms race" in data-center spending will ultimately affect their returns.
A Changing Market Landscape
Investors need to stay vigilant as they ponder whether AI-related spending patterns will persist or turn southward, potentially leading an eventual correction in markets tied heavily to these tech companies.
