AI Is Everywhere—How Investors Can Actually Diversify

Even investors who think they are diversified likely have AI exposure woven throughout their portfolios.

That is the warning from Apollo Global Management Chief Economist Torsten Slok, who argues that "a portfolio can look diversified across sectors and asset classes but still be exposed to the same underlying factor: AI."

The data backs him up. In July 2025, hyperscale data center companies accounted for just 2.7% of the U.S. investment-grade index. That share has since risen to 4.8%, and Apollo estimates it could approach 10% by 2030 as tech giants borrow heavily to fund data centers and chips.

APO stock is on the rise. Check out the charts and price action.

Historically, technology, utilities, and real estate have moved on different fundamentals, with investment-grade credit, high-yield bonds, and equities each trending up or down independently during stress periods.

Now, Slok says, AI is increasingly the thread tying these assets together—meaning a mix of stocks, bonds, and sectors that looks diversified on paper may actually be one big bet on a single theme.

Apollo's own research puts it bluntly: "Surface-level diversification across issuers and industries increasingly [masks] a single macro bet on AI," which is why the firm is steering clients toward exposures it believes are structurally insulated from AI buildout.

How to Dodge AI

Slok cites three safe havens:

European private credit. Returns here are tied to European corporate cash flows and regional credit cycles—not U.S. hyperscaler capex or Nvidia's order book.

Sports-related financing. Team and league-linked debt is backed by media rights, ticket sales, and franchise economics, cash flows that have nothing to do with GPU demand.

Hybrid credit. Structures blending debt and equity-like features, whose performance depends on deal-specific terms rather than the AI cycle.

Slok's argument is that as AI issuance balloons to roughly half of net new investment-grade supply, true diversification means actively seeking return drivers uncorrelated with the AI trade—not just spreading capital across more tickers.

Alternative Asset Managers

That read points to alternative managers positioned to sell such products.

Apollo Global Management, Blackstone, Ares Management, and KKR are all expanding private credit and sports finance platforms, positioning themselves as the go-to institutions for investors seeking to reduce AI concentration risk without sacrificing returns.

The bigger question hanging over the market: If everyone starts chasing the same "non-AI" trades at once, does that just become the next crowded bet?