Meta Invests Billions in Superintelligence AI

Mark Zuckerberg, CEO of Meta Platforms, stated on Monday that the company will invest hundreds of billions of dollars to build multiple massive AI data centers dedicated to superintelligence. This move intensifies the race for top engineering talent as major tech companies compete to hire the best minds to develop AI systems capable of surpassing human capabilities in numerous tasks.

Meta is one of the major companies that has made high-profile deals in recent months, offering multi-million dollar salaries to attract talent, with the aim of accelerating the development of machines that can surpass human thinking in many tasks. This rush underscores the importance of AI in these companies' future growth strategies.

Massive Data Centers on the Horizon

Zuckerberg reported via Threads that Meta's first multi-gigawatt data center, codenamed "Prometheus," is expected to be operational in 2026. Another data center called "Hyperion" is also being developed and can be expanded to 5 gigawatts in the coming years.

Zuckerberg added, "We are also building more superclusters, one of which alone covers a significant portion of Manhattan." This illustrates the massive scale of Meta's investments in AI infrastructure.

He also cited a report from SemiAnalysis, an industry publication, stating that Meta is expected to be the first AI lab in the world to launch a supercluster exceeding 1 gigawatt. This emphasizes Meta's ambitions to be a leader in the field of AI.

Justifying the Massive Investment

Facing investor concerns about the expected return on this huge expenditure, Zuckerberg emphasized the strong performance of the company's core advertising business, providing justification for the massive investment. He added, "We have the money from our business to do this."

The company's revenue last year was nearly $165 billion. Last month, following negative feedback about the open-source Llama 4 model and the departure of key employees, Meta reorganized its AI business into a "Superintelligence Labs" division.

Meta is betting that this division can generate new cash flows through the Meta AI app, image-to-video advertising tools, and smart glasses.

Shift in AI Strategy

The New York Times also reported that senior officials in the division are considering abandoning the company's most powerful open-source AI model, Behemoth, and shifting to developing a closed-source alternative. This potential shift indicates a change in Meta's strategy towards more controlled and potentially profitable AI models.

D.A. Davidson analyst Gil Luria said that Meta is significantly increasing its investment in AI because AI technology has already helped the company increase the number of ads and improve ad prices. He added, "But on this scale, this investment is more geared towards long-term competition, aiming to build a leading AI model, and it may take time to pay off."

In recent weeks, Zuckerberg has personally led a recruitment drive for talent at Meta's Superintelligence Labs, which will be led by former Scale AI CEO Alexandr Wang and former GitHub CEO Nat Friedman. Meta had already invested $14.3 billion in Scale.

In addition, in April, Meta raised its capital expenditure forecast for 2025 to $64 billion to $72 billion, aiming to strengthen the company's position in the competition with OpenAI and Google. The company's stock rose 1% on Monday and is up more than 20% so far this year.

Additional Analysis: Meta's focus on building a robust AI infrastructure reflects its belief that superintelligence AI will be a key driver of future growth. Investing in massive data centers and top-tier engineering talent positions Meta at the forefront of the AI race, but it also presents significant risks. The success of this strategy depends on Meta's ability to convert its AI investments into profitable products and services. The development of advanced AI technologies also raises ethical considerations regarding bias, privacy, and potential misuse, which Meta will need to address proactively.