BWX Technologies, Inc. (BWXT) Drops Over 10% Amid DeepSeek’s AI Impact and Concerns Over Nuclear Energy’s Role in AI Data Centers

In This Article:

We recently published an article titled These 29 AI Electricity, Infrastructure Stocks Are Crashing Due to DeepSeek News. In this article, we are going to take a look at where BWX Technologies, Inc. (NYSE:BWXT) stands against the other AI stocks.

Investors are pulling back from the artificial intelligence trade. Previously, a report by the Lawrence Berkeley National Laboratory highlighted that US data centers are expected to use 6.7% to 12% of all power by 2028. However, one artificial intelligence startup has upended these estimates, leaving investors wondering whether the anticipated surge in power demand and data center expansion still holds.

Energy, infrastructure, and real-estate stocks were tanking on Monday, even though they were known to be less crowded alternatives to stocks such as Nvidia. Monday’s broad-market selloff has revealed how a vast number of energy-related companies have been banking on the AI boom and the anticipated power surge it was expected to bring.

READ NOW: 10 AI Stocks Making Waves on Wall Street and 10 AI Stocks to Watch Amid the DeepSeek Buzz

“The share price drop yesterday demonstrated that many energy infrastructure companies got carried away in the momentum of the AI story last year”.

DeepSeek, an artificial intelligence startup from China, caused a frenzy in the AI world after launching its latest AI models. The company claims that these models built are at par or better than industry-leading models in the United States. They require fewer chips and are made at a fraction of the cost. All of these updates are now threatening to upset the technology world. Once the best-performing securities over the past 18 months, US electricity providers are now one of the hardest hit sectors with investors reevaluating their outlooks toward artificial intelligence and the magnitude of money that they are spending.

While many analysts, such as those at Citigroup, assert that DeepSeek’s large-language model had “prompted investor inquiries around the cost of compute”. Yet many others such as Bernstein doubt that the DeepSeek model was actually built with the acclaimed $6 million figure. They deem that the market reaction on Monday was “overblown”. Nevertheless, investors are concerned that the wider adoption of models, such as those by DeepSeek, could result in lesser demand for electricity and also require a smaller power build-out.

"If proven true, the efficiencies used within DeepSeek's open-source model can be applied by the hyperscalers to their models, which would result in a more moderated demand”.