[Meta](https://www.meta.com) is reshaping its capital structure to fuel an AI spending spree that shows no signs of slowing. The company revealed Wednesday in its [third-quarter 2026 earnings report](https://investor.atmeta.com/investor-news/press-release-details/2026/Meta-Reports-Third-Quarter-2026-Results/default.aspx) that it’s issuing a new class of non-voting stock, with shares expected to begin trading in 2027. The proceeds will bankroll AI infrastructure buildouts, potential acquisitions, and employee compensation packages as the company races to compete in what it calls an “AI transformation taking place in the industry.”
“The scale of investment required to compete, sustain growth, expand infrastructure, and pursue new strategic initiatives amidst the AI transformation taking place in the industry is significant,” Meta stated in its press release. CFO Susan Li elaborated during the earnings call that the new share class will “expand our funding tools” – corporate speak for giving the company more ways to raise cash without taking on debt or issuing traditional equity that could threaten Mark Zuckerberg’s iron grip on the company.
The strategy isn’t exactly novel. [Alphabet](https://abc.xyz) pioneered the dual-class share structure in tech when Google went public in 2004, creating non-voting Class A shares alongside voting Class B stock that kept founders Larry Page and Sergey Brin in control. Meta’s move follows the same playbook, but the timing reveals just how capital-intensive the AI race has become. While [Google](https://www.google.com) and [Microsoft](https://www.microsoft.com) have been pouring billions into AI infrastructure for years, Meta’s decision to fundamentally alter its capital structure underscores the financial pressure even the most profitable tech companies face.
The announcement comes as [AI infrastructure spending has reached unprecedented levels across the industry](https://www.techbuzz.ai/articles/google-cloud-rolls-out-nvidia-s-vera-rubin-ai-infrastructure). [OpenAI recently pushed past $50 billion in annualized revenue](https://www.techbuzz.ai/articles/anthropic-launches-claude-duet-ai-agent-to-take-on-meta-and-openai), while competitors like [Anthropic](https://www.anthropic.com) are launching increasingly sophisticated AI agents that require massive compute resources. Meta’s own AI ambitions span everything from generative AI features in Instagram and Facebook to its Llama large language models and the infrastructure needed to train them.
But there’s a catch for investors. Non-voting shares mean shareholders get economic exposure to Meta’s performance without any say in how the company is run. That’s a trade-off some investors have historically accepted from high-growth tech companies, but it also insulates management from activist pressure or takeover attempts. For Zuckerberg, who already controls Meta through a supervoting share structure, the new class adds another layer of flexibility to pursue long-term bets without quarterly scrutiny.
The timing also suggests Meta sees acquisition opportunities on the horizon. The press release explicitly mentions “potential acquisitions” as a use of proceeds, and the AI landscape is littered with promising startups that might not survive the capital requirements of scaling. [Meta has been particularly aggressive in AI talent acquisition and infrastructure buildouts](https://www.techbuzz.ai/articles/meta-scrubs-ai-training-language-after-privacy-backlash), and a dedicated war chest of non-voting shares gives the company currency for deals without impacting Zuckerberg’s control.
Employee compensation is the third pillar of the strategy, and it’s not hard to see why. AI engineers command eye-watering salaries, and stock-based compensation has been the primary tool tech companies use to compete for talent without destroying their cash flow statements. By issuing non-voting shares to employees, Meta can keep pace with [OpenAI](https://www.openai.com), [Google](https://www.google.com), and [Microsoft](https://www.microsoft.com) in the talent wars while preserving voting control.
The broader context matters too. Meta’s AI spending has already drawn investor skepticism in previous quarters, with analysts questioning whether the returns justify the capital intensity. [Infrastructure delays and power constraints have hammered other data center operators](https://www.techbuzz.ai/articles/nextdc-shares-crater-40-on-earnings-cut-power-delays), and Meta’s own Reality Labs division continues to burn billions on metaverse bets that haven’t paid off. The new share class gives Meta financial flexibility to keep spending on AI even if traditional equity markets lose patience.
Wall Street’s reaction will hinge on the details Meta provides as the 2027 launch approaches. How many shares will be issued? At what valuation? And most importantly, will the AI investments actually generate returns that justify the dilution? Those questions won’t be answered until Meta files more detailed disclosures with the SEC in coming months.
For now, the message is clear: Meta is going all-in on AI, and it’s willing to reshape its entire capital structure to do it. Whether that confidence proves visionary or reckless will depend on whether the company can turn massive infrastructure spending into products that actually make money. The clock starts ticking in 2027.
Meta’s new share class represents more than financial engineering – it’s a bet that AI infrastructure spending will define the next decade of tech competition. By giving itself a dedicated funding mechanism that doesn’t threaten Zuckerberg’s control, Meta is signaling it plans to spend aggressively on AI regardless of short-term market sentiment. Investors get exposure to that upside, but they’ll have zero say in how the money gets spent. For a company that’s already burned tens of billions on metaverse dreams, that’s a leap of faith some shareholders may not be willing to take. The real test comes when those shares start trading in 2027 and the market decides whether Meta’s AI ambitions are worth the dilution.
