Wall Street Weighs AI Slowdown Risk as Data Center Stocks Sell Off

A weekend proposal from Anthropic CEO Dario Amodei to slow the pace of frontier AI model development sent shares of infrastructure, energy, and cloud companies lower on Monday, forcing investors to reassess the durability of the data center buildout that has defined the AI era.

GE Vernova's stock dropped nearly 9% on Monday, while Caterpillar fell more than 4%, Vertiv declined close to 8%, and Oracle slipped almost 4% — a broad-based selloff reflecting how deeply the AI infrastructure thesis has been priced into industrial equities.

Amodei wrote that a slowdown in AI development does not mean "halting model training or technical progress," though that clarification did little to stem the day's losses across the sector.

The data center boom in the United States has been anchored largely to demand from two companies — Anthropic and OpenAI — and the expectation of continued, high-volume consumption of their models and services. For the constellation of companies supplying compute, energy, and physical infrastructure to that ecosystem, the calculus depends on that demand remaining uninterrupted.

Industrial heavyweights such as GE Vernova, Caterpillar, and Vertiv have committed heavily to providing the energy equipment required to power AI servers. Neoclouds including Nebius and CoreWeave have positioned themselves as critical intermediaries supplying hyperscalers with compute access. Server manufacturers Dell and Hewlett Packard Enterprise also have significant exposure to sustained AI infrastructure spending.

"Any significant delay could cost them," one tech investor told CNBC, asking not to be named in order to speak candidly.

Oracle, which has undergone a substantial transformation over the past 18 months to orient itself toward AI — laying off workers in slower-growth divisions while directing billions into compute and data centers — faces particular exposure. Rishi Jaluria, an equity analyst at RBC Capital Markets, said a slowdown in model development and training is likely a headwind for Oracle's cloud infrastructure business, which has been the primary driver of the stock's recent performance.

Even as sentiment shifted, debt activity in the sector showed few signs of retreat. A rush to secure AI-related debt financing is expected over the next six weeks, according to two people familiar with the situation who asked not to be named because the discussions are private.

Amazon returned to the debt markets last week, raising roughly £4.25 billion — nearly $6 billion — in a recent bond offering. Alphabet also tapped the European market in May, raising approximately $10 billion through a euro bond sale. One source said new debt deals announced this fall are expected to carry significantly higher interest rates than prior issuances. "Fixed income investors are demanding more reward," the person said.

The selloff also intersected with a separate analysis from Oppenheimer, which argued on Tuesday that shifting consumer behavior around AI-powered shopping could benefit platforms like Etsy. Analyst Jason Helfstein noted that consumers are relying less on traditional search engines for product discovery, with Adobe Analytics data showing that nearly half of U.S. consumers used AI tools for online shopping in June — a dynamic Oppenheimer believes will drive gross merchandise sales growth of roughly 5% over the next two years for the handmade-goods marketplace.

The divergence underscores a broader pattern emerging in AI-era markets: companies that sit at the consumer application layer may prove more insulated from slowdown fears than those supplying raw infrastructure, where the economics depend on relentless capacity expansion.

How Amodei's proposal — and the market's reaction to it — shapes capital allocation decisions in AI infrastructure over the coming months will likely determine whether Monday's selloff was a brief correction or the beginning of a longer reassessment.

Disclaimer