How the 'AI Put' has become the only thing that matters for stocks

By Joseph Adinolfi

The fate of the hard-charging bull run now hinges squarely on investors' confidence in AI, BofA says

Investors are worried, but they might be watching the wrong risk.

While investors worry a spike in global bond yields will trigger a selloff in U.S. equities, strategists at BofA Global Research say the real threat lies elsewhere.

Rising bond yields can create problems for companies looking to borrow money, and consumers angling to buy a home. But when it comes to the stock market, the trajectory of interest rates is of less concern to investors. Instead, the fate of this hard-charging bull market - which will reach its fourth birthday in October - depends entirely on what BofA is calling the 'AI put.'

Over the past few decades, when markets have hit the rocks, central bankers have reliably ridden to the rescue. On Wall Street, this dynamic has its own nickname - the "Fed put." But as the Fed embarks on another round of interest-rate hikes, a powerful momentum trade driven by investors' fear of missing out on the revolutionary promise of artificial intelligence is now accomplishing essentially the same function.

"Macro risks are struggling to compete for attention as AI FOMO dominates and drives aggressive dip-buying," the BofA team wrote in a report shared with MarketWatch.

Their thesis certainly comports with recent market action.

As yields climbed to their highest levels in decades, a previously broad-based stock-market rally was turned on its head. Stocks of small-capitalization companies and midcaps, which tend to be more sensitive to rising interest rates, turned lower. Cyclical stocks like financials and defensive sectors like utilities have both taken a hit. The Dow Jones Industrial Average DJIA, which isn't as heavily exposed to the AI theme as other popular U.S. indexes like the S&P 500 SPX and Nasdaq composite COMP, has also struggled.

At the same time, strong gains for AI-linked stocks have helped to mask much of this pain. The 20 top-performing stocks in the S&P 500 - a smattering of hot AI stocks, mostly in the tech and industrials sectors - have contributed $1.7 trillion to the market capitalization of the S&P 500 since Aug. 31, according to a Dow Jones Market Data analysis, while the bottom 480 stocks have shed about $1.9 trillion in value. The S&P 500 was on track to finish September flat on Wednesday.

The belief that AI can supercharge economic growth and corporate profits, while potentially helping the U.S. dig itself out of its fiscal hole, is helping to shield stocks from volatility in the bond market and elsewhere. As a result, the biggest threat to cross-asset stability is not high inflation, or tighter monetary policy, but the risk that the AI put might falter.

AI risks embedded in tech-analyst silos

While the Fed put was dependent on the policy deliberations of one of the world's most powerful monetary institutions, the AI put hinges on something more ephemeral - the confidence of millions of investors.

Many of them have been kicking the tires on the AI trade for years now, highlighting potential risks that could someday metastasize into bigger problems for the market. One of the most obvious problems: Where is the return on all of this AI investment going to come from?

One recent example arrived via Apollo economist Torsten Slok. Wall Street analysts generally work in silos, meaning that forecasts for profits and cash flows are often inconsistent between analysts in different sectors, even those at the same firm. But in commentary shared with MarketWatch, Slok said out that the gap between tech analysts' rosy cash-flow forecasts and the more modest numbers being penciled in for other sectors was looking suspiciously wide.

Analysts at Goldman Sachs Group and elsewhere have calculated that more than $1 trillion has already been spent on the data center build-out since the launch of ChatGPT in late 2022.

Presumably, other large American businesses would need to pay for AI tools to justify all of this investment. That money needs to come from somewhere. But while tech analysts on Wall Street expect cash flows to explode in 2028, that same optimism isn't reflected by analysts covering other sectors.

"In other words, the tech silo is betting on a future in which demand for AI and tech services explodes, while the silos covering the companies that would pay for those services see a much more modest outlook," Slok said.

The BofA team acknowledged that there probably is a level at which rising Treasury yields could start to weigh on stocks - it is just higher than most in the market probably expect.

And with so much hinging on the AI put, any loss of confidence could quickly collide with an uncertain macroeconomic backdrop, helping magnify losses across markets.

Ken Jimenez contributed

-Joseph Adinolfi

This content was created by MarketWatch, which is operated by Dow Jones & Co. MarketWatch is published independently from Dow Jones Newswires and The Wall Street Journal.

(END) Dow Jones Newswires

09-30-26 1659ET