Some of the world's biggest investors are quietly reducing their exposure to US stocks, worried that too much of their money is riding on a single theme: artificial intelligence. But this isn't a wholesale retreat. A new survey of 430 large institutions found that a third plan to cut their US stock holdings over the next year — double the share that said the same a year ago. The move is less about predicting a crash and more about managing risk.
Why concentration is the worry
The concern boils down to how heavily the US stock market is weighted toward a handful of giant companies. The ten biggest stocks in the S&P 500 now make up roughly 40% of the index, well above the 30-year average of about 25%. And virtually all of those top names are tied to AI in some way — from chipmakers to cloud-computing giants.
That concentration reaches far beyond professional money managers. A typical global index fund invests about 64% of its money in US stocks, which means anyone holding such a fund is leaning on AI more than they might realize. For everyday investors, this is a useful reminder that a diversified fund can still carry a heavy tilt toward one sector.
Not a crash call, just a hedge
The professionals aren't calling for a market collapse. Instead, they're saying they don't want so much riding on one theme. Trimming a position is different from dumping it. Big funds want to keep some exposure to AI's upside, but they also want a cushion if the trade unwinds.
This is a classic risk-management move. When a single theme drives a large share of returns, even investors who believe in the long-term story often reduce their bets to avoid being caught off guard by a sudden reversal.
Where else can investors look?
Buying overseas stocks seems like an obvious fix, but AI's influence has spread globally. An Asian economic watchdog warned this week that the region is especially exposed to an AI slump. Japanese and Hong Kong stocks now move in tandem with US tech, and chipmaking heavyweights dominate some Asian benchmarks. Samsung and SK Hynix together make up half of South Korea's Kospi index, while TSMC accounts for at least 40% of Taiwan's main index.
Bonds aren't a clean escape either. AI-linked companies have become the biggest borrowers in high-quality US corporate debt, meaning even bond investors are indirectly tied to the sector's fortunes.
Some institutions are looking off the beaten path. Hedge funds bought European stocks in September at the fastest pace in over five years, drawn by a market with relatively few AI giants. India is also gaining fans as a large market with little direct exposure to AI. But these alternatives come with their own trade-offs — spreading bets has often meant lagging behind while AI stocks keep climbing.
What it means for everyday investors
For the average person, this story is less about predicting the next market move and more about understanding what you already own. If you hold a global index fund, you're likely more exposed to AI than you think. That's not necessarily a problem — AI has been a powerful driver of returns — but it's worth being aware of.
Diversification doesn't mean avoiding AI altogether; it means not letting any single theme dominate your portfolio. The big institutions are trimming, not fleeing, because they still see value in the technology. The same logic can apply to individual investors: you don't have to sell everything, but it may be wise to check how much of your portfolio is tied to one sector.
As always, the key is to stay informed and make decisions based on your own goals and risk tolerance. For more on how valuations and earnings are interacting, see our piece on high stock valuations meeting strong earnings. And if you're curious about how AI companies are faring in the bond market, our analysis of AI borrowers facing higher junk-bond yields offers a deeper look.
For a broader view of global markets, you might also check European stocks rebounding and the latest on the dollar holding steady.
