Wall Street has treated AI infrastructure stocks as one interchangeable trade for most of this buildout cycle, rising and falling together on the same demand headlines.
The Monday, Sept. 14, sell-off broke that pattern. Bernstein did not just flag sector-wide risk from a potential AI slowdown. It named the single stock that stood to lose the most.
The warning traces back to the weekend. Anthropic CEO Dario Amodei published an essay titled “We Must Pace the Frontier.” In it, he argued that AI companies need to slow down how they improve model capabilities.
He warned that unchecked progress, particularly as AI systems begin to self-improve, is outstripping the industry’s ability to safely test and control the technology. OpenAI CEO Sam Altman and Elon Musk both backed the call within hours.
The timing complicates Amodei’s message. Anthropic confidentially filed for an IPO in June and is widely expected to list within weeks, according to CNBC. It means the same executive urging restraint is also asking investors to fund faster growth.
The contradiction unsettled a market that had priced years of uninterrupted spending into stock valuations. CoreWeave Inc. (CRWV), the cloud provider that rents AI computing capacity to labs and enterprises, was Bernstein’s answer. Its shares fell about 8% on Sept. 14, one of the sector’s steepest declines, according to Seeking Alpha.
The market’s reaction to these concerns remained stark through the end of the trading session. CoreWeave shares failed to recover from their early morning slide, closing down 6.75% at $82.98.
The drop reflects how exposed investors now see CoreWeave’s business model if the industry slows down.
Related: David Tepper makes surprising double bet on AI’s biggest bottleneck
Why CoreWeave carries more risk than its peers
Bernstein analyst Madison Rezaei said the exposure comes down to geography. About 25% of CoreWeave’s active U.S. power sits in Tier 3 and Tier 4 markets, Seeking Alpha reports.
Those are rural sites built for training, not the low-latency inference work that dominates elsewhere. Another 74% of its power that is contracted but not yet active sits in those same markets, according to Bernstein.
That matters because training tolerates distance from users, while inference does not. A slowdown in training would hit demand for power that CoreWeave has not yet locked into contracts, Rezaei said. Its existing backlog is largely protected by take-or-pay terms, so that risk is more contained.
Bernstein rates CoreWeave Underperform with a $74 price target, implying meaningful downside from current levels. Wall Street overall remains more constructive, with an average price target near $139, according to TipRanks.
The rural buildout is the flip side of a capital-spending ramp already underway. The company raised its 2026 forecast to $35 billion to $39 billion, according to its own second-quarter results. The same report showed its revenue backlog swelled past $104 billion over the same stretch.
That spending funded much of the latency-insensitive capacity Bernstein now flags as vulnerable, and it came with real leverage. CoreWeave carried roughly $35 billion in debt as of the end of June, the same report showed.
Metro-based data centers look safer to Bernstein
Equinix, Digital Realty and Csquare sit on the other side of Bernstein’s split. Each keeps more than 90% of its U.S. capacity in major or minor metro markets, at 95%, 92% and 94%, respectively, according to Seeking Alpha.
Bernstein rates all three Outperform, with price targets of $1,270 for Equinix, $226 for Digital Realty and $27 for Csquare.
More AI:
- Nvidia just made a move Wall Street wasn’t ready for
- Microsoft just took sides in AI policy fight
- OpenAI just disclosed something genuinely alarming
Metro locations are key because inference, the work that runs AI products after training, needs to sit close to users and customers.
Bernstein called those markets the safest and most valuable part of the sector as it shifts toward inference.
The sell-off reached far beyond data centers
The Sept. 14 decline was not confined to CoreWeave or its direct rivals. It spread across the companies that supply and connect AI infrastructure.
- Neocloud peers dropped alongside CoreWeave, with Nebius down 8%, Applied Digital down 6%, DigitalOcean down 5%, and IREN down 4%, according to Seeking Alpha.
- Networking equipment makers fell sharply, with Lumentum down 8%, Ciena and Coherent each down 7%, and Arista Networks down 5%, the same report showed.
- Chipmakers tied to the AI buildout slid, too, as Marvell and Arm each fell 8%, AMD and Intel dropped 6%, and Broadcom fell 4%.
The AI trade just stopped being one basket
For most of this cycle, infrastructure stocks moved together, rising and falling on demand headlines rather than balance sheets or site maps.
Bernstein’s note marks a shift toward the opposite instinct: pricing where a company’s capacity physically sits, not just how much of it exists.
That distinction will matter more if Amodei’s call for deliberate pacing gains traction. “We must slow the pace at which we improve the capabilities of AI models,” he wrote.
A slower training cycle would not erase AI demand. It would reallocate that demand toward inference, and inference rewards proximity to customers over raw scale.
The pattern echoes the telecom fiber buildout of the late 1990s. Demand for bandwidth was real, but timing and location errors sank operators years before traffic caught up to the capacity.
Investors who spent the past two years buying AI infrastructure as a single trade may now need a map, not just a thesis.
