On Monday, there was quite a stir in the U.S. stock markets. The reason for this sudden panic was speculation surrounding a company called DeepSeek and its artificial intelligence (AI) models. Some investors were worried that DeepSeek’s advancements in AI could significantly decrease the demand for graphics processing units (GPUs), which are crucial pieces of hardware used extensively in AI and gaming. As a result, Nvidia, a major manufacturer of GPUs, saw its stock price fall nearly 20%, indicating that investors were concerned about the potential impact of DeepSeek’s technology.
Despite this market reaction, Meta, the parent company of Facebook, is not backing down. During a financial call on Wednesday, CEO Mark Zuckerberg made it clear that Meta is committed to investing heavily in AI technology. He described this investment as something that could involve “hundreds of billions of dollars” over time, which signals that the company sees AI as vital for its future.
Read Also: SoftBank Discusses Investing Up To $25B In OpenAI, Report Claims
Just a week prior, Zuckerberg had announced that Meta plans to spend over $60 billion in 2025 alone on various infrastructure projects, the majority of which will be used to build and upgrade data centers. These facilities are essential for storing and processing vast amounts of data needed for AI operations.
When asked about how DeepSeek could affect Meta’s spending in AI, Zuckerberg emphasized that investing in AI infrastructure is essential for maintaining a competitive edge. He referred to DeepSeek as a new rival, but mentioned that it’s still too soon to determine whether the demand for chips (like GPUs) will decline, as they are still incredibly important for executing AI tasks, especially given Meta’s large user base, which consists of billions.
Zuckerberg expressed his belief that building a strong AI infrastructure would provide Meta with significant advantages in both the quality of its services and its ability to manage the scale of operations that they are aiming for.
Looking ahead, Meta is working on a new AI model called Llama 4. Zuckerberg stated that the goal with this model is to make it the best in the world, competing even against proprietary models like ChatGPT. He mentioned that Llama 4 is expected to have “agentic capabilities,” which means it will be capable of performing tasks independently in a way that is similar to what companies like OpenAI and Anthropic are developing. Additionally, he noted that this model will also support multiple forms of information (like images and text), which is referred to as “multimodal.”
In summary, while market reactions to competition can be intense—like the drop in Nvidia’s stock this week—Meta remains resolute in its commitment to AI, planning significant investments for the future. They are determined to create cutting-edge AI models that can compete at the highest level, maintaining a strong position in a rapidly evolving technology landscape.
