A New Funding Model for a New AI Era
The move comes as Big Tech faces the astronomical cost of building and maintaining the data centers needed to power generative AI systems. Historically, companies like Meta, Amazon, and Google have relied almost entirely on internal capital to expand their infrastructure. But with the scale and speed of AI advancements demanding more flexibility and financing, that model is evolving.
“We’re exploring ways to work with financial partners to co-develop data centers,” Meta Chief Financial Officer Susan Li said during the company’s post-earnings call on Wednesday. While internal funding remains the primary source for Meta’s capital expenditure, Li noted that some future projects may rely on “significant external financing,” providing optionality if infrastructure requirements change.
Though no specific deals have been finalized, Meta’s latest financial disclosure suggests plans are advancing. The company stated that it approved a plan in June to dispose of certain data center assets, reclassifying $2.04 billion worth of land and construction-in-progress as "held-for-sale." These assets are expected to be transferred to third-party partners within the next 12 months to support joint data center developments.
As of June 30, total assets classified as held-for-sale had climbed to $3.26 billion. Meta emphasized that the reclassification did not result in any losses, with valuations based on the lower of carrying amounts or fair market value less costs to sell.
Zuckerberg’s Billion-Dollar AI Vision
At the heart of this funding strategy is Meta’s ambitious push into AI. CEO Mark Zuckerberg has laid out plans to build vast data center “superclusters” designed to support the company’s long-term vision for artificial general intelligence (AGI).
“Just one of these covers a significant part of the footprint of Manhattan,” Zuckerberg said, underscoring the sheer scale of the infrastructure Meta is building.
Reflecting this ambition, the company this week raised the lower end of its annual capital expenditures forecast by $2 billion, revising the range to $66 billion to $72 billion. The increased spending will go toward expanding its AI capabilities, from custom silicon to training infrastructure and the deployment of large-scale language models across its products.
Ad Revenue Keeps the AI Engine Running
Despite the steep capital outlays, Meta’s core business remains healthy. The company reported stronger-than-expected advertising revenues, fueled by AI-driven enhancements to ad targeting and content recommendations on platforms like Facebook, Instagram, and WhatsApp.
Executives credited these AI improvements for helping offset some of the rising infrastructure costs, suggesting that the same technology driving the company’s next phase of growth is also bolstering its current financial performance.
Looking Ahead
Meta declined to provide additional details on its data center asset strategy, but the disclosure marks a notable departure from the company's historical approach to infrastructure investment. As the cost of AI development continues to soar, the move could foreshadow a broader trend among tech giants seeking outside capital to stay competitive in the AI arms race.
With hundreds of billions of dollars in future investments on the horizon, Meta’s willingness to share the burden may be less a retreat than a calculated repositioning—one that allows the company to scale more efficiently while keeping its eyes on the ultimate prize: AI dominance.