The AI hedge portfolio just bled 10% in five days. The high-beta momentum basket? Down 12%. Headlines scream 'AI bubble bursting.' Goldman Sachs says otherwise. Their latest note doesn't declare the trade over. It declares the phase over. The era of buying any ticker with 'AI' in the description is finished. What replaces it is a surgical, fundamentals-driven rotation. And the first target is hiding in plain sight: storage and data centers.
This isn't a call to abandon the sector. It's a call to re-enter it with a scalpel instead of a sledgehammer. The market is repricing AI assets from 'potential' to 'proof.' The next leg up belongs to companies with earnings, not just narratives.
The Context: From Beta to Alpha
For two years, the AI trade was a rising tide. Semiconductors, cloud providers, even adjacent software—everything floated. The strategy was simple: buy the basket, ride the wave. That worked until it didn't. The recent deleveraging is a violent correction of that lazy positioning. Leverage built during the euphoria phase is being unwound. The 'AI trade' as a monolithic bet is over.
Goldman's data confirms this shift. Momentum factors are rebalancing. Software has now overtaken semiconductors as the largest weight in the three-month momentum long portfolio. Simultaneously, semiconductors and the broader AI complex have moved into the short basket. This is a massive signal. The market is no longer rewarding chipmakers for simply existing. It's demanding they prove demand. Meanwhile, software—the application layer—is gaining favor. The bet is shifting from 'who builds the brain' to 'who uses it effectively.'
This is the classic transition from infrastructure build-out to application monetization. The picks and shovels phase is maturing. The next phase rewards those who can turn the raw compute into revenue.
The Core: Storage and Data Centers Are the New Frontline
Goldman's specific recommendation cuts through the noise: storage and data centers. The logic is pure quantitative structural skepticism. The valuation gap is the widest in the market. The profit recovery in this segment hasn't been fully priced into the stock prices. This is a classic dislocation between price and fundamentals.
Let's break down the mechanics. The AI build-out requires massive capital expenditure. That capex flows first to GPUs—Nvidia's domain. But the bottleneck is shifting. You can't run AI workloads without data storage. You can't house the servers without data centers. The memory and storage companies (think Micron, Western Digital) and the infrastructure providers (Dell, Super Micro) are the downstream beneficiaries of the same capex wave. The market has been so fixated on the GPU that it's ignored the plumbing.
My own analysis of the supply chain confirms this. The lead times for high-bandwidth memory (HBM) are stretching. Data center leasing rates are climbing. The demand is real. The earnings, however, haven't caught up to the stock prices of the chip designers. This creates the arbitrage Goldman is pointing to. The profit recovery is the catalyst. The stock price is the lagging indicator.
This is a tactical play, not a strategic one. It's a 3-6 month window to capture the repricing. The risk is that the 'profit recovery' is already priced in. But the data suggests otherwise. The market is still punishing anything associated with the AI complex indiscriminately. That's the opportunity.
The Contrarian Angle: The Copper Signal and the Second Wave
Here's what the mainstream analysis misses. Goldman mentions capital flowing to 'neglected areas'—European and Japanese banks, gold miners, and copper stocks. This isn't just a flight to safety. It's a recognition of a new AI-driven demand curve. AI data centers are power-hungry. They require massive electrical infrastructure. That means copper. Lots of it. The market is starting to price in the physical resource requirements of the AI build-out. This is the 'picks and shovels' trade for the next phase.
The contrarian view is that the AI trade isn't ending. It's bifurcating. The first wave was about the compute layer. The second wave is about the physical layer—power, cooling, storage, and the raw materials to build it all. The money flowing to copper miners isn't a retreat from AI. It's a bet on AI's physical footprint.
But there's a darker side to this rotation. The deleveraging may not be over. The AI hedge portfolio's 10% drop is a warning shot. If Nvidia's Q2 earnings disappoint, or if the September industry conferences fail to deliver a new catalyst, we could see a second, more violent wave of selling. The storage and data center trade would not be immune. It would be dragged down with the broader complex. The window is open, but it's not without risk.
The Takeaway: The Catalyst is the Calendar
This is a data-driven, event-driven market. The next major pivot points are clear: Nvidia's Q2 earnings and the September industry conferences. These events will either validate the current rotation or trigger a second leg of the deleveraging. The smart play is to position for the rotation but respect the risk. Don't buy the entire AI basket. Buy the specific segments where the earnings gap is widest. Watch the copper and power infrastructure names. They are the new leading indicators.
The AI trade isn't dead. It's just getting smarter. Speed is the only currency that doesn't inflate. The market is moving fast. The question is whether you're positioned for the rotation or still holding the bag from the last phase. The data is clear. The playbook is written. The only variable left is execution.