Meta launches $35bn stock offerings to fund AI infrastructure

Structure of the offerings

Meta has launched two stock offerings totalling $35bn to finance its AI infrastructure and capital projects. The first offering involves $22.5bn of Class A common stock through forward sale agreements, which Meta will settle by 19 November 2027. The second offering includes $12.5bn of mandatory convertible preferred stock, set to convert into common stock around 15 August 2029 and pay dividends until then.

Morgan Stanley and Goldman Sachs are managing both sales. Underwriters have the option to purchase up to $3.375bn of additional Class A common stock and 37.5 million more preferred shares within 30 days of the offering.

Interest from chip suppliers

Nvidia, AMD, and Broadcom have expressed non-binding interest in purchasing up to $5bn of the Class A common stock. This interest aligns with Meta’s capital expenditure plans, largely directed toward AI-related hardware, including data centres and semiconductors.

Use of funds and financial context

Meta will use the proceeds for general corporate purposes, primarily funding its $135bn to $145bn capital expenditure budget for 2026. Most of this budget is allocated to data centres and chip infrastructure, supporting AI initiatives like the Muse agent, a product launched in September.

The company’s free cash flow turned negative in the third quarter of 2026, though it generated $97bn in cash from operations in the first nine months of the year. Meta held $82bn in cash and securities at the end of September, providing a financial cushion alongside the new funding.

Future share sale plans

Meta has also announced plans for a separate $60bn programme to sell nonvoting Class C shares on the open market, beginning in early 2027. No agreement for this programme has been finalised yet.

Market reaction

Meta’s stock initially fell as much as 4% following the announcement before recovering to trade about 1% higher.

Link to AI spending trends

This funding move follows earlier reporting on Meta’s push into AI-driven products, such as the Ray-Ban Meta smart glasses and the Muse agent. The company’s capital expenditure has increasingly shifted toward AI infrastructure, including partnerships like its $35bn deal with CoreWeave for computing capacity. This aligns with broader industry trends, where tech giants are scaling investments in AI hardware and data centres to support long-term growth.

What this means

Lazyfounder analysis — our interpretation, not reported fact.

Meta’s $35bn stock offerings signal a significant escalation in its AI infrastructure investments. For founders and operators, this underscores the scale of capital required to compete in AI, particularly in hardware and data centre capacity. The involvement of chip suppliers like Nvidia, AMD, and Broadcom suggests deepening partnerships between tech giants and semiconductor manufacturers, which could shape pricing, supply chains, and innovation cycles in the AI hardware market.

For startups, Meta’s spending trajectory highlights the importance of strategic capital allocation. While most startups lack Meta’s resources, the focus on long-term AI infrastructure—rather than short-term profitability—demonstrates how large players are prioritising future growth over immediate returns. This could create opportunities for smaller players specialising in niche AI applications or cost-efficient hardware solutions, but it also raises the barrier to entry for competitors aiming to match Meta’s scale.

Related on Lazyfounder

• Meta
• Goldman Sachs
• Nvidia
• Broadcom
• CoreWeave
• More finance news
• Ray-Ban Meta glasses add privacy setting to block AI training data storage