Meta announced on Monday that it plans to raise a total of $95 billion in equity to bankroll its investments in AI infrastructure and superintelligence, one of the largest capital raises ever attempted and a decisive answer to the question of how the social media giant will pay for its compute ambitions. Shares rose 1.6 percent in premarket trading after the announcement, according to CNBC.
The structure comes in two parts. Meta is initially launching underwritten public offerings of roughly $35 billion in common stock and convertible preferred stock, and it will then raise up to an additional $60 billion through an at-the-market program, which sells shares into the open market over time. In a statement, Meta said Nvidia, AMD and Broadcom have expressed interest in purchasing up to $5 billion in shares — a detail that underscores how tightly the AI industry's suppliers and customers have become financially entangled. You can track this and the latest AI developments as the raise executes over the coming months.
Breaking Down the $95 Billion
Meta said the proceeds will support what it calls emerging AI opportunities, including its family of apps, its Muse AI agent, and its growing data center business. The company plans to balance funding across a mix of operating cash flow, investment-grade debt, partnerships, and equity capital, meaning the equity raise is one pillar of a broader financing strategy rather than a standalone move.
The chipmakers' involvement is notable. If Nvidia, AMD and Broadcom buy into the offering, a portion of Meta's infrastructure spending will effectively flow back to the companies selling it GPUs and networking gear — the same circular financing pattern that has drawn scrutiny across the industry, from Nvidia's investments in AI labs to chip vendor credit lines for data center projects.
Why Meta Needs the Money
The raise follows a third-quarter earnings report last week that showed both Meta's strength and its strain. Revenue jumped 26 percent year-on-year to $64.60 billion, beating the $63.37 billion analysts expected, per CNBC. But investors focused on the company's negative free cash flow and a higher full-year capital expenditure forecast — Meta now expects to spend $135 billion to $145 billion on capex this year, up from a bottom-of-range estimate of $130 billion. The stock fell on that news at the time.
The physical build-out behind those numbers is staggering. Meta says it is constructing four "superclusters," each with more than a gigawatt of capacity, and that it is on track to nearly double its total infrastructure capacity in 2026 compared with the year before. Equity financing eases the pressure on a balance sheet that has already leaned heavily on debt: the company's bond issuances and off-balance-sheet data center structures have drawn increasing attention from investors worried about where the cash will come from if AI revenue lags spending.
Chipmakers Buy In — and the Industry Races to Raise
The chipmakers' interest in the offering fits a broader pattern. Nvidia has taken stakes in AI labs and infrastructure companies; AMD has committed to purchasing billions in Anthropic stock as part of its GPU supply agreement; and Broadcom has financed Anthropic through convertible notes. Suppliers funding their own customers has become a defining feature of the AI boom's financial architecture.
Meta is also far from alone in going to the market. Alphabet said in June it would raise $85 billion. SpaceX closed a $40 billion debt financing to purchase AI chips, as we covered earlier today. Anthropic filed its S-1 for a Nasdaq IPO the same morning as Meta's announcement. CNBC reported Friday that TikTok U.S. is seeking multiple gigawatts of computing capacity in the coming years. "Meta offering more stock comes at a time when numerous companies, such as Google, SpaceX and soon, Anthropic, are going to the markets to raise funds," Ben Wood, senior partner at CCS Insight, told CNBC, describing "the eye-watering levels of investment required to keep up in the AI arms race."
The AI Capex Supercycle
The scale of Meta's raise is easier to absorb against industry-wide forecasts. Goldman Sachs estimates that AI-related capital spending could total about $7.6 trillion from 2026 through 2031 across compute, data centers, and power, according to CNBC. Against that backdrop, a $95 billion equity program is less an outlier than a marker of what hyperscale AI competition now costs.
The market's reaction also marks a turnaround from June, when Meta shares sank more than 5 percent after the Financial Times reported the company was considering raising tens of billions for its AI push. Investors initially balked at the idea of dilution; on Monday they bid the stock up, a signal that the market has largely accepted massive external financing as the cost of admission to the superintelligence race.
What to Watch
Three questions will determine whether the raise is remembered as shrewd or excessive. First, execution: the $35 billion underwritten portion needs to clear at a healthy price, and the $60 billion at-the-market component will drip into the market for months, testing demand continuously. Second, dilution: $95 billion in new shares is a material claim on future earnings, and Meta has not said exactly when the ATM sales will run. Third, payoff: Meta is betting that Muse, its apps, and its data center business can convert gigawatts into revenue fast enough to justify capex that exceeds its free cash flow.
For now, the message is straightforward. Meta has decided that staying at the front of the AI race is worth asking shareholders to help pay for — and its chip suppliers, it seems, are happy to buy in.
Stay Ahead of AI
The AI build-out is accelerating — track every development as it happens.
Read more AI news →
