Meta Platforms on Tuesday raised the planned size of its largest-ever equity offering to $97.5 billion, as the social media giant turns to the public market to aggressively expand its artificial intelligence infrastructure. The move marks a significant escalation in the company’s fundraising ambitions and underscores just how expensive the race to build frontier AI systems has become. For a business that spent years funding its ambitions from the cash generated by its advertising machine, the decision to ask public investors for tens of billions of dollars represents a notable philosophical shift, one that reflects both the scale of the opportunity Meta perceives and the sheer capital intensity of pursuing it.
The fundraising plan, among the biggest ever mounted by an AI company, has also drawn interest of up to $5 billion in total from Nvidia, Advanced Micro Devices and Broadcom for Class A common shares. That funding interest adds to concerns about an increasingly circular flow of capital in the AI industry, as chip suppliers help finance the expansion of one of their biggest customers. Meta has been pouring money into chips and data centers to compete better with OpenAI and Anthropic, and the participation of its own suppliers in the offering raises uncomfortable questions about where the boundary between customer and investor now sits. When the companies selling the picks and shovels also buy shares in the miners, the financial relationships become difficult to untangle, and the risk of overstated demand becomes harder for outsiders to assess.
Meta now plans to raise $25 billion in common stock at $625 per share, compared with $22.5 billion disclosed on Monday. It also priced 250 million convertible preferred depositary shares at $50 apiece, for a total offering value of $12.5 billion. The offerings include about $37.5 billion of common and convertible preferred stock. The Facebook parent’s shares closed down nearly 2 percent at $630.56 on Monday, and the stock was down about 1 percent on Tuesday, a muted reaction that suggests investors are weighing the dilution against the strategic logic of the spending.
The social media company also has a separate program to sell up to $60 billion in shares over time at prevailing market prices. While Meta has not yet finalized the timing and terms for its at-the-market offering, it expects the sale to begin after its planned non-voting Class C shares start trading in the first quarter of 2027. On Tuesday, Nvidia and AMD separately announced investments of $3 billion and $1 billion, respectively, in Meta’s equity offering. Those commitments carry symbolic weight as much as financial weight, signaling that the semiconductor firms most exposed to Meta’s spending plans have a vested interest in seeing them succeed.
William Blair analysts offered a measured assessment on Monday. “We expect investors to continue to question the financing and potential for future capital raises,” they said. “Overall, we believe the financing reinforces both Meta’s confidence in the long-term returns from AI investment and the increasingly capital-intensive nature of pursuing that opportunity.” That framing captures the tension at the heart of the transaction. Confidence and capital intensity are two sides of the same coin here, and the analysts are effectively acknowledging that the market will need repeated reassurance before it fully accepts the story.
Meta’s latest fundraising further underscores a shift among companies increasingly tapping debt markets and raising equity to bolster their AI infrastructure, after previously relying on cash to fund their investments. Alphabet had increased its equity offerings to $84.75 billion in June. The scale of these moves suggests that the largest technology firms no longer view internal cash flow as sufficient to keep pace with the capital demands of AI development, and that investors are being asked to shoulder a growing share of the risk. The pattern is now well established across the sector: hyperscalers are layering equity raises, convertible notes, and corporate debt on top of record operating profits, a strategy that would have seemed reckless a decade ago but has become the norm in an era defined by compute scarcity.
There is a practical dimension to all of this that deserves attention. Data centers capable of training and serving frontier models require land, power, cooling, and networking at a scale that few utilities or municipalities are prepared for, and the timelines for bringing such facilities online are measured in years rather than quarters. The capital being raised now is effectively a bet on demand that will materialize later, and if the pace of AI adoption slows or if efficiency gains reduce the need for raw compute, the returns on these investments could disappoint. Meta’s leadership has been explicit about its willingness to accept that risk, but shareholders are the ones ultimately carrying it.
The convertible preferred structure also merits a closer look. By pricing depositary shares at $50 apiece with conversion features, Meta is appealing to investors who want income alongside equity exposure, a segment of the market that has grown considerably in recent years. The structure gives the company flexibility, but it also introduces complexity into the capital stack that will need to be managed carefully as the Class C shares begin trading in early 2027. How those instruments are received will offer an early signal about investor patience with the broader AI spending cycle.
What remains genuinely open is whether the market’s appetite for this kind of financing holds. The largest technology firms are now competing not just on models and products but on their ability to convince investors that the enormous sums being committed will eventually produce commensurate returns. Meta’s offering is a test of that proposition, and the participation of Nvidia, AMD, and Broadcom adds a layer of mutual dependency that cuts both ways. If the AI buildout delivers, the arrangement will look prescient. If it stalls, the circularity will be cited as a warning sign that was visible all along. For now, the company has secured the capital it sought, and the judgment on whether it was wisely deployed will take years to render.
