Chinese AI Startup Leads U.S. Data Center, Energy Stocks To Plunge
Shares of U.S. tech, data center and energy firms plummeted Monday morning following the unveiling of DeepSeek, a Chinese artificial intelligence model that rivals those of U.S. tech giants — and was produced at a fraction of the cost while using far less computing power.
The emergence of DeepSeek’s R1 model has raised questions about the hundreds of billions of dollars in investments that tech giants like Microsoft, Amazon, Google and Meta are making on high-performance AI chips, data centers and the energy infrastructure needed to support them.
While the best models from U.S. tech firms still outperform R1, the efficiency of the Chinese model reportedly had Meta engineers “in panic mode” and has generated amazement from tech leaders like venture capitalist Marc Andreessen, who wrote on X that it is “one of the most amazing and impressive breakthroughs I’ve ever seen” and AI's “Sputnik moment.”
The model's $5.6M price tag and development on fewer, more primitive chips have suddenly cast doubt on a fundamental assumption underpinning Big Tech’s AI spending spree: that massive amounts of energy-intensive chips are needed to develop more advanced AI models that would lead to commercially viable products.
DeepSeek’s success has raised the possibility that this thesis may be flawed, and that it will take far fewer chips — and potentially far fewer data centers and far less power — to achieve these aims.
AI hardware giant Nvidia took the biggest hit in Monday's selloff, with shares dropping close to 17% by midday and remaining at that level until the market closed.
The Big Tech behemoths driving the AI boom also fell sharply. Oracle and Softbank, who jointly announced the $500B Stargate data center project last week, were down by roughly 14% and 10%, respectively at the end of the day Monday.
With DeepSeek’s efficiency injecting sudden uncertainty into the data center sector’s previously robust demand picture, data center REITs Equinix and Digital Realty, which trades as DLR, also saw their share prices drop by 4.3% and 8.7%, respectively.
“There may be a knee-jerk negative reaction (more so on DLR) to the news, given fears of a likely rethink of big AI capex spend,” Mizuho analyst Vikram Malhotra wrote in a note to clients, according to SeekingAlpha.
He added that if the claims about R1’s efficiency are proven true, there could be a drop in Big Tech capex that he says would be “a clear near-term negative for data center REITs developing wholesale centers and stocks that have big growth premiums built in.”
Wall Street’s fears of diminished demand for data center capacity are also impacting the energy sector, where growth projections for electricity and natural gas demand were largely predicated on growing consumption from AI data centers. As a result, shares of firms like Constellation Energy and GE Vernova fell sharply in Monday's trading.
“The question becomes if you can do so much more, so much more efficiently, do you then need as much data centers, do you need the spend?” CNBC host David Faber said shortly after the markets opened Monday morning. “That’s why Constellation and GE Vernova, which has run up so much, are getting hit today.”
Still, some leaders within the U.S. tech sector are putting a more positive spin on DeepSeek’s advances. In a post on X, Microsoft CEO Satya Nadella suggested that cheaper AI will speed up adoption of the technology and accelerate the expansion of the AI market as a whole.
“As AI gets more efficient and accessible, we will see its use skyrocket, turning it into a commodity we just can't get enough of.”
Other analysts have expressed doubt that DeepSeek will have a significant impact on the data center demand landscape. Wedbush's Dan Ives told The Wall Street Journal that he believes Big Tech will continue to spend on data centers, projecting $2T in capital spending over the next three years. Additionally, he expressed doubt that American firms would ever use a Chinese firm for their AI infrastructure.
Both in Silicon Valley and Wall Street, there is also some skepticism around whether DeepSeek’s model is as efficient as the company claims or whether additional computing may have been used in R1’s development. But others have labeled the ongoing uncertainty as a wakeup call for investors about the volatility of an AI ecosystem that, despite hundreds of billions of dollars in investment, is still in its infancy.
“The problem is that the AI industry is embryonic,” wrote a Deutsche Bank analyst, according to the WSJ. “And it’s almost impossible to know how it will develop or what competition current winners might face even if you fully believe in its potential to drive future productivity. The stratospheric rise of DeepSeek reminds us of this.”
UPDATE, JAN. 27, 4:50 P.M. ET: This story has been updated to reflect the stock prices when the market closed Monday.
