Markets Stocks Economy Crypto Earnings Banking Energy
Home› Markets› Feature
Markets · Exclusive

Wall Street scrambles to map hidden AI risk after selloff

Wall Street scrambles to map hidden AI risk after selloff
Markets · 2026
Photo · Eleanor Whitfield for Daily Digest Invest
By Eleanor Whitfield Markets Editor-in-Chief Nov 16, 2026 5 min read

After a dizzying run-up and a recent pullback in AI-related stocks, Wall Street is doing something less glamorous but arguably more important: figuring out exactly how much of their portfolios are tied to the technology. Portfolio managers who once treated artificial intelligence as one of the market's biggest trades are now running new screens to map both the obvious and the hidden risks embedded across stocks and bonds.

The exercise is partly defensive. When a trade gets as crowded as AI has been, the danger isn't just that the underlying companies disappoint—it's that everyone tries to exit at once. That's why managers are now paying close attention to what they call "spillover exposure": the ways AI shows up in places you wouldn't immediately think of, from a trucking firm using software to cut fuel costs to a municipal bond whose tax base depends on a new data center.

The direct and the indirect

The "direct" bucket is the easiest to spot. According to MSCI's Global Intelligence Index, about $25.3 trillion is invested in the largest global companies with the clearest links to AI—the chipmakers, cloud providers, and software giants that dominate the headlines. That's a staggering number, and it explains why a wobble in a single AI bellwether can move entire indices.

But the trickier part is the indirect exposure. AI doesn't just live in the companies that build it; it lives in the companies that use it, supply it, or are financed by it. A logistics firm that uses AI to optimize routes is still an AI beneficiary, even if its ticker doesn't scream "tech." A city that issues bonds to fund a data center is now tied to AI's fortunes, even if the bond is rated investment-grade.

That's why PGIM, an asset manager overseeing nearly $1 trillion in fixed-income strategies, has started running every bond and loan it holds through AI screens. The goal: judge how the technology could change borrowers' cash flows—for better or worse—and test what happens if AI-related spending cools or a shock hits an unexpected corner of the market.

Why the sudden urgency?

The urgency comes from the scale of the exposure. Equable Institute estimates that a basket of 52 AI-linked public companies made up 38.5% of public equity holdings at a group of large pension funds as of June 2026. When a single theme accounts for more than a third of a portfolio, the line between "investing in AI" and "betting the farm on AI" starts to blur.

That concentration has a knock-on effect. If a few big, benchmark-aware investors cut their AI exposure to stay within risk mandates, others often follow—not because they think AI is doomed, but because they don't want their portfolios to drift too far from an index. This is what's known as tracking error, and it can force even reluctant managers to sell.

Nomura Asset Management's Andy Goldberg warns that even a small reduction—say, 1% to 2%—can look like a rout if everyone tries to exit the same crowded trade at once. The biggest air pockets, he says, may show up in "AI beneficiaries": the indirect, theme-adjacent names that can be less liquid than the mega-caps that dominate major indices. In a selloff, those are the stocks that can gap down hardest when there are few buyers.

What it means for investors

For the average investor, this is a reminder that market risk isn't always where it appears. If you own a broad index fund, you already have AI exposure—whether you asked for it or not. And if you own individual stocks, it's worth asking how much of their revenue depends on AI spending, either directly or through their customers.

The good news is that the market is becoming more aware of these hidden links. The fact that managers are running AI screens on bonds and loans is a sign that the industry is trying to get ahead of the next shock, rather than react to it. That's a healthy development, even if it means some short-term volatility as positions get adjusted.

For now, the key takeaway is simple: AI is no longer just a tech story. It's a market-wide story, and the risks are spreading into corners that once seemed safe. As Nvidia's CEO probe shakes Wall Street's top picks, the focus on hidden exposure is likely to intensify. And with softer producer prices lifting Wall Street's mood, the broader market may be in a forgiving mood—but that doesn't mean the risks have disappeared.

Investors should keep an eye on how much of their own portfolios is tied to AI, and remember that the most crowded trades can unwind in unexpected ways. The next few months will likely test whether the market's newfound caution is enough to prevent a repeat of the recent selloff—or whether the hidden exposure is still bigger than anyone thinks.

More from this story

Next article · Don't miss

TPG in talks to take BILL Holdings private, Reuters reports

TPG is in talks to take BILL Holdings private, potentially at a premium, after activist investors pushed for changes. The news sent BILL shares up nearly 13% to a 52-week high.

Read the story →
TPG in talks to take BILL Holdings private, Reuters reports