Meta Outlines Massive AI Infrastructure Financing

Meta is placing the scale of its artificial intelligence ambitions in financial focus, outlining a $97.5 billion funding plan for the computing infrastructure behind its next generation of products. The proposal includes substantial equity offerings, but not every part of the headline figure has been secured.

The company has described approximately $37.5 billion in underwritten public offerings, divided between about $25 billion in Class A common shares and $12.5 billion in depositary shares representing preferred stock. A further programme of up to $60 billion could allow stock to be sold into the market over time. Meta has cautioned that this additional programme may never be established.

Chipmakers are also part of the proposed financing. Nvidia has committed $3 billion and AMD $1 billion, while Nvidia, AMD and Broadcom could together purchase up to $5 billion in shares. Their involvement highlights the commercial ties between Meta and the suppliers of the processors needed to develop increasingly demanding AI systems.

The spending plan accompanies a substantial physical build-out. Meta expects capital expenditure of $135 billion to $145 billion in 2026 and is developing as many as four data centre superclusters, each exceeding one gigawatt of capacity. Those facilities represent long-term requirements for electricity, cooling, equipment and reliable network connections.

For the infrastructure industry, the announcement illustrates how AI expansion is changing the scale of data centre development. Large computing campuses cannot be delivered through chip purchases alone. Grid access, construction capacity and operational efficiency are becoming decisive factors, while investors must distinguish commitments already underwritten from financing that remains conditional.

Meta reported third-quarter revenue of $64.6 billion and net income of $9.9 billion, providing a measure of its resources alongside the size of its ambitions. Those profits do not, however, remove the need to assess the cost of maintaining such vast facilities. The proposed financing may accelerate construction, but its larger significance rests in how effectively Meta can translate expensive computing capacity into durable demand.