Meta plans non-voting stock class to fund AI expansion
Meta plans to create a new class of non-voting capital stock as it expands investment in artificial intelligence. Chief financial officer Susan Li confirmed the proposal during the company’s third-quarter earnings call, saying the shares will be distributed to all existing shareholders through a stock dividend.
The new stock is intended to create a publicly traded non-voting instrument that Meta can use for capital raising, employee compensation, acquisitions and other general corporate purposes. Li said the structure would preserve the voting power held by existing Class A and Class B shareholders and maintain Meta’s founder-led structure.
A funding tool alongside higher AI spending
Meta reported $64 billion in third-quarter revenue, a 26% increase from a year earlier, beating Wall Street expectations. At the same time, its costs and expenses increased 68% year on year, underscoring the scale of the company’s investment commitments as it develops and deploys AI products.
Li said the new shares would expand Meta’s funding tools and provide flexibility to invest in momentum across its business that the company attributes to AI. The plan does not alter the voting rights of the existing Class A and B share classes, but it would give Meta a distinct security for future corporate uses.
Reality Labs remains a material cost centre. The unit, which develops Meta’s AR glasses, VR headsets and VR software, is still losing money, although Meta expects its losses to decline from 2025 levels to between $18 billion and $19 billion.
Muse adoption is part of the AI case
Meta has recently released several AI models and products, including Muse Spark, Muse Glimmer, Muse Wonder, the open-weight Muse Spark Open model and the Muse Code coding agent. Its development of AI coding products follows Meta’s agentic coding push with Muse Spark 1.1 as Meta broadens its tools for software development and other AI use cases.
At Meta Connect, the company also introduced Meta Muse, a line of avatars for its AI assistant. Mark Zuckerberg said that more than 10 million people in the United States and Canada had set up Muse personal agents in a month and a half. He described the product as having clear product-market fit and said Meta was entering a more operational phase focused on scaling it.
Legal scrutiny remains in view
Li also noted that active legal and regulatory matters could significantly affect Meta’s business and financial results. She cited youth-related scrutiny in several markets and a number of youth-related trials scheduled in the United States in the coming months.
For businesses tracking Meta, the practical implication is that AI product adoption and revenue growth should be considered alongside capital requirements, rising expenses, governance protections and regulatory exposure when evaluating the company’s ability to scale its AI programme.