Meta Platforms is exploring the creation of a new class of stock as a mechanism to fund its sweeping artificial intelligence ambitions, according to a report from TechCrunch. The move would represent a significant structural shift in how the company finances its technology buildout, with AI infrastructure sitting at the center of the rationale.
To understand why this matters, it helps to appreciate the scale of what Meta has committed to on the AI front. The company has pledged to spend tens of billions of dollars building out data centers, acquiring chips, and developing its own large language models — including the Llama family of open-weight models that it has released publicly. That spending has already drawn scrutiny from investors who remember the bruising years of the metaverse bet, when capital poured into Reality Labs with modest returns to show for it. CEO Mark Zuckerberg has essentially asked shareholders to trust him twice in a decade on transformative technology bets. The AI chapter is the second ask, and it is a much larger one in dollar terms.
Creating a new class of stock is not, in itself, an unusual corporate maneuver. Alphabet has multiple share classes, as does Meta itself — Zuckerberg already holds Class B shares that carry disproportionate voting power, a structure that insulates him from shareholder pressure when making long-horizon bets. What makes this reported move notable is the specific purpose attached to it: financing AI expenditure. The likely reading is that Meta wants a way to raise capital or compensate stakeholders — employees, partners, or both — through equity instruments that are structurally separated from the existing share classes, potentially without diluting Zuckerberg's voting control or triggering the same kind of market reaction that a straight equity offering might produce.
The broader industry pattern here is worth pausing on. The major technology companies are all grappling with the same fundamental tension: AI infrastructure is extraordinarily expensive to build, the returns are not yet fully proven at scale, and equity markets are watching the burn rate closely. Microsoft has leaned on its partnership with OpenAI to share some of that exposure. Google has the advantage of owning the full stack from chips to cloud to consumer products. Amazon is building out its own models while selling infrastructure capacity to others. Meta's position is unusual because it does not have a hyperscale cloud business in the traditional sense to cross-subsidize the AI spend. Its revenue is almost entirely advertising, which means every dollar of AI capital expenditure comes out of cash flows that Wall Street already has strong opinions about.
A new share class, if structured cleverly, could give Meta a currency that lets it pursue AI deals, attract talent, or secure partnerships without touching the core capital structure in ways that alarm existing investors. It might also signal that the company expects its AI assets to eventually carry independent value — a way of internally ring-fencing what could become a separately monetizable business.
The consequences of this move, should it proceed, would ripple in several directions. For existing Meta shareholders, the critical question is dilution and governance. If the new shares carry no voting rights and are issued in ways that fund AI buildout without weakening the balance sheet recklessly, many institutional investors may accept the logic. If, however, the structure is seen as a mechanism to entrench management further while writing a blank check to an AI strategy that lacks a clear near-term revenue model, the reaction could be considerably cooler. Activist pressure on big tech has been relatively muted compared to earlier cycles, but it has not disappeared.
For the broader AI industry, a Meta move of this kind would likely accelerate a trend of treating AI infrastructure as a discrete financial category rather than simply a line item in capital expenditure. It could prompt competitors and analysts to ask whether the major AI spenders should be valued differently — more like infrastructure companies or utilities than advertising platforms.
For regulators and antitrust observers, complex share class structures at companies already under scrutiny for market power are always worth noting, even if there is nothing inherently problematic about the mechanism itself.
The things worth watching in the coming weeks and months are fairly clear. First, whether Meta files any documentation with the Securities and Exchange Commission that formalizes this new class, which would reveal the actual structure and its intended uses. Second, how large institutional shareholders — the asset managers and pension funds that hold significant Meta positions — respond publicly or through their proxy advisory apparatus. Third, whether any of this capital architecture is linked to specific AI product milestones, which would at least give the market a framework for judging whether the spending is producing results. The AI buildout at Meta is not slowing down. The question is simply who pays for it, and on what terms.
