Zuck shrugs off DeepSeek, vows to spend hundreds of billions on AI

American markets panicked on Monday due to speculation that Deepseek’s AI models would crush the demand for GPUs, with the shares of Nvidia falling nearly 20%.

But Meta does not deteriorate, with his CEO Mark Zuckerberg promised that the tech giant would invest “very heavy” in AI even “hundreds of billions of dollars”-hey in the long term during Meta’s win in the first quarter of Meta On Wednesday.

Zuckerberg already announced last week that Meta would spend more than $ 60 billion on capital expenditure in 2025 alone, mainly on data centers.

In response to the question of an analyst about the impact of Deepseek on Meta’s AI expenditure, Zuckerberg said that spending on AI infrastructure will remain a “strategic advantage” for Meta.

Meta regards Deepseek as a new competitor and learns it, but it is “way too early” to say whether the demand for chips will stop because they remain crucial for inference purposes, Zuckerberg said and noticed that Meta has billions of users.

“At the moment I would bet that the ability to expand that type of infrastructure will be a big advantage for both the quality of the service and the scale we want,” said Zuckerberg.

The goal of Meta with its next model, Llama 4, is to make it the most competitive in the world, even in comparison with closed models (such as Chatgpt), Zuckerberg said. He added that he expects it to have agental possibilities – something that both OpenAi and Anthropic have been moved – together with Multimodale.

“Our goal with Llama 3 was to make open source competitive with closed models,” he said. “And our goal for Lama 4 is to lead.”

WAN has an AI-oriented newsletter! Register here to get it in your inbox every Wednesday.