Story
Oct 27, 2026
Meta’s $95 Billion AI Push Pulls Nvidia and AMD Into Its Orbit
Meta sees a route to “personal superintelligence” for billions, while its chip partners see a chance to embed their hardware in the next generation of AI services. For investors, the same plan is a reminder that the AI race is becoming as much a financing contest as a technology contest.
On Monday, Meta laid out the financial scale of its AI ambitions: a planned $95 billion stock raise to fund infrastructure and superintelligence efforts. The company launched roughly $35 billion in underwritten equity offerings and said it intends to add a $60 billion at-the-market programme once its Class C shares begin trading in early 2027.1
Meta framed the capital call as an offensive move, arguing that AI is opening opportunities across its established products and new services. Its stated goal is to expand computing capacity and bring “personal superintelligence to billions of people and businesses.”1 Yet the move also underscores the strain of the build-out: Meta expects annual capital expenditure of $135 billion to $145 billion, even as rising costs have squeezed operating income.2
The wider market is confronting the same bill. Goldman Sachs estimates AI-related spending across compute, data centres and power could reach $7.6 trillion between 2026 and 2031.2 CCS Insight’s Ben Wood called Meta’s fundraising a reflection of the “eye-watering levels of investment required to keep up in the AI arms race.”2
By Tuesday, two of Meta’s crucial suppliers had moved from prospective buyers to committed investors. Nvidia said it would put $3 billion into the offering; AMD followed with $1 billion.3 The money comes with commercial logic: Nvidia and Meta will deepen work on Muse, Meta’s AI-agent platform, while Meta expands deployments of Nvidia’s Vera Rubin systems and Vera CPUs.3
Nvidia chief Jensen Huang cast the partnership as a shift in how people use technology: “AI agents represent a new computing paradigm — software that understands our goals, reasons, uses tools and gets work done on our behalf.”3 Zuckerberg, in turn, said Meta was eager to work with Nvidia on “the best personal agents and frontier models for billions of people.”3
AMD offered the infrastructure counterpart, saying its Helios racks and EPYC processors are already running at Meta. Its chief executive, Lisa Su, said the companies were building systems at “gigawatt scale” for Meta’s most ambitious models and services.3 The wager is clear: Meta gets capital and silicon; its chipmakers gain a deeper stake in the AI demand they are helping create.