Morgan Stanley has signaled that the era of AI investors pouring money almost exclusively into semiconductor stocks may be coming to an end, with the investment bank recommending a rotation toward hyperscalers — the tech giants spending billions on AI infrastructure — as well as consumer discretionary, transport, and biotechnology shares.
In a note published Monday, Morgan Stanley said the recent weakness in U.S. semiconductor stocks is a sign that the market gains are broadening, suggesting that the next phase of the AI cycle will benefit a wider range of companies beyond the chipmakers that have dominated Wall Street for the past two years.
The end of the chip-only trade
The Philadelphia SE Semiconductor index climbed 11% in June alone, but has fallen over 11% in the two weeks prior to the note — a pullback that Morgan Stanley interprets not as the end of the AI boom but as a normal rotation.
While the likes of Nvidia and AMD have seen their valuations soar to historic heights on the back of insatiable demand for AI training chips, the note argues that clear evidence that AI products can generate returns that justify the massive spending is yet to be seen. This uncertainty is pushing investors to look beyond the chip sector.
Hyperscalers poised to benefit
Morgan Stanley identified hyperscalers — including Alphabet, Amazon, Meta Platforms, and Microsoft — as primary beneficiaries of the rotation. These companies have committed billions to scale up their AI infrastructure, and their stocks have already gone through a period of underperformance in June.
Hyperscalers are expected to drive about 40% of total Russell 1000 cash capital expenditure over 2026-2028, representing more than $2 trillion in spending on AI data centers, networking, and computing infrastructure. As AI shifts from training to inference — actually running AI models in production — the value is expected to flow increasingly to the companies operating the infrastructure rather than just those building the chips.
Broader market rotation
Beyond hyperscalers, Morgan Stanley pointed to consumer discretionary goods, transportation, and biotechnology as sectors that could benefit from the rotation out of the red-hot chips trade. The brokerage noted that markets paring back expectations of rate hikes by the U.S. Federal Reserve, along with a fall in crude oil prices, is also driving this shift.
Meanwhile, hedge funds have already been positioning for this rotation. According to a Goldman Sachs client note, U.S. hedge funds sold tech hardware stocks for a fourth consecutive week in late June and early July, suggesting that institutional investors are already moving ahead of the trend.
What this means for AI investing
The Morgan Stanley call represents a potential inflection point for the AI investment narrative. For two years, the conventional wisdom has been that the surest way to profit from AI was to invest in the companies making the chips — Nvidia, AMD, Broadcom, and the memory manufacturers like SK Hynix and Samsung. The semiconductor index has more than doubled over that period.
But as the AI industry matures, the argument goes, the value will shift to the companies actually deploying and monetizing the technology at scale. The hyperscalers, with their massive cloud platforms and AI services, are best positioned to capture this value.
Whether the rotation materializes will depend on upcoming earnings reports from both chipmakers and hyperscalers, as well as the trajectory of AI spending. But Morgan Stanley's note suggests that the days of betting indiscriminately on any company with a chip may be numbered.
