As the AI boom moves into its next phase, investors are starting to realize that not everyone in the space is playing the same game — and that difference may matter a lot more in 2026.
After a volatile end to 2025 marked by tech sell-offs, debt raises, and eye-watering valuations, some market watchers see the turbulence as an early warning sign. The AI trade, long treated as a single rising tide, is beginning to split into distinct camps. According to Stephen Yiu, chief investment officer at Blue Whale Growth Fund, the market has yet to seriously differentiate between companies building AI, companies spending heavily on it, and those actually making money from it — but that’s likely about to change.
Yiu describes three broad groups emerging: private AI developers and startups like OpenAI and Anthropic, publicly listed companies pouring billions into AI infrastructure, and the firms on the receiving end of that spending, such as chipmakers and infrastructure suppliers. So far, investors — particularly retail investors exposed through ETFs — have largely lumped them together. “Every company seems to be winning,” Yiu noted, but he cautioned that AI is still early, and that kind of thinking won’t last.
Valuations are already reflecting the strain. Many of the so-called Magnificent Seven are trading at hefty premiums after ramping up AI investment, even as free cash flow gets squeezed by massive capital expenditures. Blue Whale, which focuses on free cash flow yield to judge valuation, has grown wary of companies burning cash to fund AI ambitions without clear returns. Yiu says his preference is to sit “on the receiving end” of AI spending rather than backing the biggest spenders themselves.
Other strategists echo the need for sharper distinctions. Barclays’ Julien Lafargue says the froth isn’t spread evenly across the market, but concentrated in areas where optimism has raced ahead of earnings — pointing to pockets like quantum-adjacent plays as particularly risky. In those cases, investor enthusiasm may be doing more of the work than fundamentals.
At the same time, AI is reshaping Big Tech’s business models. Companies once prized for being asset-light are now becoming capital-intensive hyperscalers, investing heavily in data centers, GPUs, land, and power. That shift changes how they should be valued. As Schroders’ Dorian Carrell put it, treating these firms like traditional software companies may no longer make sense when huge upfront costs and uncertain returns are baked in.
Debt has helped fund this transformation, with companies like Meta and Amazon tapping credit markets, though their strong balance sheets offer some reassurance. Still, the bigger test is coming. If new AI revenues fail to outpace rising costs, margins will come under pressure — and depreciation on expensive infrastructure will start showing up in the numbers.
Taken together, these trends point to a more fragmented AI market ahead. Instead of one broad AI trade, investors may soon be forced to choose between builders, spenders, and beneficiaries — each with very different risk profiles. As Yiu put it, the real story going forward isn’t whether AI will change the world, but who actually captures the value when it does. And in 2026, that differentiation is likely to become impossible to ignore.
This analysis is based on reporting from CNBC.
Image Courtesy of Unsplash.
This article was generated with AI assistance and reviewed for accuracy and quality.
