Meta CEO Mark Zuckerberg said on Wednesday (Jan. 29) that it’s still too soon to say how DeepSeek’s breakthrough AI model will affect his company or the tech industry as a whole.
That said, he pushed back on the idea that Meta will start spending less on AI anytime soon despite claims that DeepSeek has shown cutting-edge AI can be built with far less money and computing power than expected.
Zuckerberg stated: “It’s probably too early to really have a strong opinion on what this means for the trajectory around infrastructure and CapEx.
“There are a bunch of trends that are happening here all at once.”
However, it appears that Meta is continuing its trajectory of expanding its AI capabilities. During Meta’s first-quarter earnings call, Zuckerberg pledged that the tech giant would “very heavily” invest “hundreds of billions of dollars” in AI over the long term.
Meta reported a solid 21% jump in sales year over year, hitting $48.39 billion, which beat Wall Street’s expectations.
On Monday (Jan. 27), Meta’s shares rose to nearly 2%, suggesting that some investors think DeepSeek’s cost-efficient AI breakthrough could help the company build powerful AI systems for less money.
Meta plans to invest billions to expand AI
Meanwhile, last week, Meta announced plans to invest between $60 billion and $65 billion this year as part of its AI expansion. And on Wednesday, the company said it expects total expenses for 2025 to land between $114 billion and $119 billion.
Zuckerberg said Meta will keep learning from technologies like DeepSeek’s while continuing to release its own open-source and free Llama AI models to help drive the industry forward.
He also pointed out that DeepSeek’s emergence reinforces the company’s belief in an open-source approach to AI. He said: “There’s going to be an open-source standard globally.
“For our own national advantage, it’s important that it’s an American standard.”
Featured image: Grok