Meta Creates Non-Voting Shares to Fund AI

Meta has confirmed plans to issue a new class of non-voting stock to help finance its aggressive expansion into artificial intelligence. This move allows the company to raise significant capital for its AI initiatives while preserving the voting control held by its existing shareholders. The announcement was made during the company's third-quarter earnings call on Wednesday.

The new shares will be distributed to all current holders as a stock dividend. Unlike traditional shares, these do not carry voting rights in company decisions. This structure is designed to maintain Meta's founder-led governance model. It provides the corporation with additional financial tools to support its rapid growth.

Susan Li, Meta's chief financial officer, explained the rationale behind the new equity. She stated that the goal is to create a publicly traded instrument for capital raising and employee compensation. The company intends to use these shares for acquisitions and other general corporate needs. Li emphasized that this approach preserves the voting power of existing Class A and B shareholders.

The financial backdrop for this decision is strong. Meta reported revenue of $64 billion for the quarter, representing a 26 percent increase from the same period last year. These results exceeded Wall Street expectations. However, the company's costs and expenses also surged, rising by 68 percent year-over-year. This spike reflects the heavy investment required to build out its AI infrastructure.

The Reality Labs division, which handles virtual reality headsets and augmented reality glasses, continues to operate at a loss. Yet, the bleeding is slowing down. The company expects this division's losses to narrow significantly. For the upcoming year, losses are projected to fall between $18 billion and $19 billion, down from 2025 levels.

Meta has been highly active in the AI space over the past year. The company released several models, including Muse Spark and Muse Glimmer earlier this year. Recently, it launched Muse Wonder and Muse Spark Open. The latter is an open-weight model, meaning its underlying parameters are publicly available for other developers to use and build upon.

In addition to large language models, Meta introduced Muse Code, a specialized agent for software developers. At its recent Meta Connect event, the company unveiled a new line of avatars for its AI assistant, Meta Muse. These characters are designed to be cute and cuddly, aiming to make the technology feel more relatable and shareable for users.

Early adoption metrics suggest these efforts are gaining traction. CEO Mark Zuckerberg noted that more than 10 million people in the U.S. and Canada have set up Muse personal agents in just six weeks. He described the growth as strong and mentioned that the product has ranked number one in the App Store nearly every day during this period.

Zuckerberg indicated that the company is entering a new operational phase focused on scaling the technology. He expressed confidence that the product has found product-market fit. The long-term goal is to expand the service to eventually serve billions of people globally. This ambition drives the need for the new funding mechanisms.

Despite the positive momentum, Meta acknowledges ongoing challenges. The company is monitoring various legal and regulatory matters that could affect its future. Li pointed to increased scrutiny regarding youth-related issues in several markets. There are several trials scheduled in the coming months in the United States that the company is actively preparing for.