Meta to pour hundreds of billions into AI infrastructure as Zuckerberg shrugs off DeepSeek threat

Meta CEO Mark Zuckerberg

Meta chief Mark Zuckerberg isn't too worried about the rise DeepSeek despite the Chinese artificial intelligence (AI) lab's quick ascent shaking up Silicon Valley and the Wall Street.

If anything, it has "strengthened our conviction that this is the right thing for us to be focused on", Zuckerberg said on January 29, referring to Meta's open-source approach with its foundational large language model (LLM) Llama.

"There's a number of novel things that they did that we're still digesting ... number of advances that we will hope to implement in our systems and that's part of the nature of how this works," Zuckerberg said during the company's earnings conference call.

The Meta boss said that every new company that has a launch, whether it is a Chinese rival or not, will have some new advances that the rest of the field learns from.

Over the past week, DeepSeek has sent Wall Street — particularly AI-related stocks — into a meltdown with its claims of building a model that can rival top-tier models from American companies such as OpenAI, Meta and Google at a fraction of the cost. This had caused concerns among investors over the billions of dollars being poured in by tech companies to develope their AI models and products.

During the earnings call, Zuckerberg said he still believes heavily investing in capital expenditures and infrastructure will be a strategic advantage over time "It's probably too early to have a strong opinion on what this means for the trajectory around infrastructure and capex," he said.

Meta will invest "hundreds of billions of dollars" in AI infrastructure over the long term, Zuckerberg said. Last week, he announced that Meta will spend $60 billion to $65 billion in capital expenditure in 2025 to bolster its AI efforts.

He said much of the compute infrastructure will likely move from the pre-training phase towards building powerful "reasoning" models and high-quality products to billions of consumers.

This "doesn’t mean you need less compute" because you can "apply more compute at inference time in order to generate a higher level of intelligence and a higher quality of service”, Zuckerberg said.

"As a company that has a strong business model to support this I think that's generally an advantage that we're now going to be able to provide a higher quality of service than others who don't necessarily have the business model to support it on a sustainable basis," he said.

Llama 4 launch in coming months

Meta plans to launch Llama 4 with natively multimodal and agentic capabities in the forthcoming months. "Llama 4 is making great progress in training. Llama 4 mini is done with pre-training and our reasoning models and larger model are looking good too," Zuckerberg said.

"Our goal with Llama 3 was to make open source competitive with closed models, and our goal for Llama 4 is to lead," he added.

Zuckerberg also said 2025 will be the year when it "becomes possible to build an AI engineering agent that has coding and problem-solving abilities of around a good mid-level engineer".

"This is going to be a profound milestone and potentially one of the most important innovations in history, as well as over time, potentially a very large market. Whichever company builds this first will have a meaningful advantage in deploying it to advance their AI research and shape the field," he said.

Zuckerberg also expects the company's AI assistant Meta AI to reach more than one billion people this year.

"We have a really exciting roadmap for this year with a unique vision focused on personalisation. We believe that people don't all want to use the same AI -- people want their AI to be personalized to their context, their interests, their personality, their culture, and how they think about the world," he said.

Financial performance

Meta, which owns Facebook, Instagram, and WhatsApp, reported a revenue of $48.4 billion for the fourth quarter of 2024, a 21 percent increase year-on-year (YoY), while its net income jumped 49 percent to $20.8 billion.

Meta's family daily active people (DAP), which are users open to any one of its apps in a day, grew 5 percent YoY during the quarter to 3.35 billion.

Reality Labs, the unit which develops the company's virtual reality and augmented reality hardware and software, posted an operating loss of $4.97 billion for the quarter, as compared to $4.7 billion loss in the same quarter last year.

Revenues were at $1.1 billion, led by sales of hardware such as the Ray-Ban Meta smart glasses and Quest VR headsets.