Meta is doing two things at once, and both are about the same problem: keeping its chief executive committed and its control structure intact. As SiliconANGLE reported, the company beat third-quarter revenue expectations and unveiled nonvoting Class C shares that will act as a two-for-two stock split. Separately, as Wired reported, Mark Zuckerberg has agreed to stay in California despite a billionaires' tax ballot measure, with a new stock agreement giving him financial incentives to stay after Meta's board worried his departure could disrupt its AI ambitions. The thread is not the AI race itself but the governance machinery Big Tech is building around it: stock as a retention tool, nonvoting shares as a way to raise capital without diluting founder control, and a founder whose personal and political circumstances now sit inside a corporate risk assessment. That is the pattern worth watching, and it says more about the state of US technology companies than any single earnings figure.
Control Is Now a Product Feature
The Meta story is often told as a founder behaving eccentrically. It is better read as a board managing a key-person risk. Wired reports that Meta's board feared Zuckerberg's departure could disrupt the company's AI ambitions, and that the new stock agreement gives him financial incentives to stay. That is a governance admission: the company's AI strategy is treated as inseparable from one person, so much so that a state tax ballot measure becomes a board-level concern. The nonvoting Class C shares, announced alongside the revenue beat, point the same way. A two-for-two split that issues nonvoting stock lets Meta raise capital and widen its shareholder base without handing new holders a say in how the company is run. For a US market that has spent years debating dual-class structures, this is the next step: not just dual class, but an expanding pool of shareholders with no vote at all.
The AI Bill Comes Due Anyway
The irony is that Meta is tightening founder control at the very moment the AI buildout is forcing it to ask public markets for capital. The SiliconANGLE report ties the nonvoting shares explicitly to raising capital. That is a company that needs money and wants to keep its hands on the wheel while taking it. The same tension appears in the IPO debate. Crunchbase News carried a guest article from Airrived co-founder and CEO Anurag Gurtu arguing that Anthropic's rapid revenue growth does not justify its proposed $2 trillion IPO valuation, given enormous infrastructure commitments, safety concerns, customer concentration and competition from cheaper open-weight models. Whatever one makes of the number, the argument is about whether AI revenue can carry the infrastructure commitments behind it. Meta is answering that question on the balance sheet, by structuring a capital raise so that the people writing the checks do not get to steer. For US technology companies, that is the trade being made across the sector: capital is welcome, control is not negotiable.
Where Enterprise AI Value Is Said to Come From
The Crunchbase News piece offers a second thread worth taking seriously: the claim that lasting enterprise AI value will come from proprietary data, reliable deployment and measurable customer outcomes rather than expensive models alone. That is a direct challenge to the model-centric valuation logic that has driven the current AI IPO wave, which the article likens to the dot-com bubble's evil twin. If that view holds, the companies that win are not necessarily the ones with the largest training runs but the ones that can point to deployment and outcomes. That reframes the Meta story too. Meta's AI ambitions are the stated reason its board wants its CEO anchored in California. The question a US investor should ask is not whether Meta can build a large model but whether the spending converts into products customers pay for and keep paying for. The governance arrangements above are, in part, a way to keep that question from being put to a shareholder vote.
