Meta CEO Mark Zuckerberg has dismissed concerns over the rapid rise of Chinese AI company DeepSeek, assuring investors that Meta remains confident in its artificial intelligence (AI) strategy. Speaking during Meta’s fourth-quarter earnings call on Wednesday, Zuckerberg addressed multiple questions from Wall Street analysts about DeepSeek’s cost-efficient AI breakthroughs and their impact on the broader AI landscape.
While DeepSeek has caused a significant sell-off in AI stocks due to fears that powerful AI models may no longer require extensive computing resources, Zuckerberg said the company’s achievements only reinforce Meta’s commitment to AI. “What DeepSeek was able to accomplish with relatively little money has only strengthened our conviction that this is the right thing to be focused on,” he said.
Despite the buzz around DeepSeek’s AI models, Zuckerberg assured investors that Meta is actively studying the company’s innovations and intends to implement some of its advancements into its own Llama AI models. “There’s a number of novel things they did we’re still digesting,” he said, adding that Llama 4, Meta’s next AI model, will feature multimodal and agentic capabilities and is expected to roll out in the coming months.
Zuckerberg also pushed back against the idea that DeepSeek’s breakthroughs will make Meta’s investment in AI infrastructure less valuable. He argued that while AI training may become more efficient, “this doesn’t mean you need less compute,” as inference—the process of running AI models—still requires significant resources. “You can apply more compute at inference time in order to generate a higher level of intelligence and a higher quality of service,” he explained.
Deepak Sharma, Co-Founder & Managing Partner at IA commented, "DeepSeek has open-sourced key components of its model, pushing AI innovation further. This move not only accelerates AI democratisation but also signals a shift in how foundational AI models are developed and shared globally. DeepSeek’s success underscores a new era where leaner, more agile startups can challenge industry leaders, reshaping the AI landscape."
Meta is set to spend nearly $60 billion on AI this year, and Zuckerberg took a veiled jab at competitors like OpenAI and Anthropic, emphasizing that Meta has a “strong business model” to sustain its AI ambitions. He suggested that some AI startups lack sustainable revenue streams, making their future uncertain.
He also highlighted Meta’s ability to scale AI across its vast user base, predicting that Meta’s AI assistant will reach one billion users this year—a significant advantage over competitors.
Beyond AI, Meta’s financial results showcased its continued dominance in the digital space. The company reported $48.39 billion in revenue for Q4 2024, up 22% from the previous year, with a net profit of $20.8 billion, a 43% increase.
Zuckerberg also took a moment to praise President Donald Trump, saying, “We now have a US administration that is proud of our leading companies, prioritises American technology winning, and will defend our values and interests abroad.”
Notably, this statement came just as reports emerged that Meta has agreed to pay Trump $25 million to settle a lawsuit over his account ban following the January 6th insurrection. Most of the settlement funds will reportedly go toward Trump’s presidential library.
Despite recent controversies and changes to its content policies, Meta’s platforms continue to grow. CFO Susan Li confirmed that ad spending remains strong, and daily active users across Meta’s apps reached 3.35 billion in Q4, a 5% increase from the previous year.