AI Rally Changes Course: Why JPMorgan Advises Looking Beyond Nvidia

Let's be honest: over the last year and a half, investment strategy in AI has looked indecently simple. You bought chip manufacturers (you know who), a few cloud giants, and the profit chart soared into the stratosphere. But in markets nothing lasts forever, and Mira Pandit, global market strategist at JPMorgan Asset Management, brought news that sobers you up: the easy walk has ended, the real work has begun.

The essence of Pandit's statement is that so-called "broadening out" of artificial intelligence trading is happening. This doesn't mean the bubble has burst. It means it's changing shape. The money that used to vacuum up exclusively hardware manufacturers is now beginning to flow into other vessels.

What changed?

The market has become saturated with promises of a bright future and is now looking for those who will serve that future physically. Investors realized a simple thing: for GPT-5 (or 6, or 7) to even work, you need not only H100 graphics processors. You need gigawatts of energy, advanced cooling systems, copper cables and real estate for data centers. This is why we're seeing a cyclical pivot toward utilities, energy and industrial sectors. This is no longer glamorous tech sector, but harsh infrastructure, without which AI will remain just code on a server.

The second important point that JPMorgan highlights is geography. For a long time, the AI rally was a purely American party. Now the focus is shifting to international markets. In Europe and Asia, there are companies integrated into supply chains or developing applied software, but costing many times less than overheated American counterparts. Smart money is looking for undervalued assets where others aren't yet looking.

Why is this important right now?

We are at a turning point. Expectations from Big Tech are inflated to the limit — any mistake in a quarterly report is punished by an instant drop in stock prices. At the same time, "boring" sectors that are winning from AI implementation (for example, companies optimizing logistics or production with neural networks) are showing growth potential. This is classic capital rotation: investors fix profits on race leaders and shift them to laggards, who now have a growth driver.

For the industry, this is a signal of maturity. Hype is turning into business processes. If before we paid for "shovels" (chips), now we're starting to pay those who dig with these shovels, and those who sell water to gold diggers.

The main point: The bet on "winner takes all" no longer works. The AI portfolio of the future is not a list of five tech giants, but a complex mix of energy, industry and international software. Are you ready for such diversification?