According to Woofun AI, Meta (META.US) exhibited a sharp divergence between strong revenue growth and negative cash flow in the third quarter of 2026. Founder Mark Zuckerberg is aggressively investing in infrastructure to compress free cash flow to -$2.12 billion, securing long-term dominance in AI strategy.
This fundamental shift in the financial structure is driven by strong growth in advertising revenue and improved efficiency in monetizing traffic. In the quarter, Meta generated total revenue of $64.601 billion, a 26% year-over-year increase compared to $51.242 billion in the same period of 2025; even after excluding a $113 million negative impact from foreign exchange, revenue growth under constant exchange rates remained at 26%. The core driver was advertising, with advertising revenue reaching $63.061 billion, up 26% from $50.082 billion in the same period last year. Other revenue from the family of apps amounted to $1.097 billion, bringing total revenue for this segment to $64.158 billion, while the Reality Labs division, focused on the metaverse and hardware, generated $443 million in revenue. On the traffic side, daily active users across the family of apps reached 3.64 billion, up 3% year-over-year.
More importantly, there has been a deepening of monetization capabilities: ad impressions increased by 10% year-over-year, and the average price per ad rose by 14%. Management attributes this performance to improvements in the AI recommendation system, believing that intelligent algorithms are significantly enhancing user engagement and ad targeting precision, thereby unlocking incremental value within a saturated traffic landscape.
However, the income statement was severely impacted by massive litigation fines, leading to a significant contraction in net profit. In the third quarter, Meta recorded a $9.89 billion settlement expense related to lawsuits concerning adolescent mental health, a substantial one-time cost that directly weighed on overall profitability. As a result, consolidated operating profit fell to $12.988 billion, a 37% year-over-year decline; reported net profit amounted to $9.86 billion, with diluted earnings per share of $3.83. Even more dramatically, on October 21—the same day the earnings report was released—a U.S. District Court judge in New Mexico, Francis Mathew, ruled in favor of the state’s lawsuit against Meta for misleading consumers about data privacy, ordering Meta to pay up to $25 billion in penalties to the state.
This severe legal penalty not only intensified immediate financial pressure but also cast a long-term shadow over the company’s future compliance costs and potential liabilities, significantly eroding what was once robust advertising profits through legal expenses.
The core of the cash flow crisis lies in the rapid surge in capital expenditures far outpacing the growth in operating cash flow, leading to negative free cash flow. Although Meta’s operating cash flow for the quarter remained high at $32.786 billion, exceeding $29.999 billion in the same period of 2025, spending increased even more sharply. Cash outflows for the purchase of property and equipment rose to $33.942 billion, compared to just $18.829 billion in the same period of 2025; when adding $964 million in principal payments on finance leases, total related cash outflows reached $34.906 billion. According to Meta’s official figures, free cash flow for the period declined to negative $2.12 billion. In contrast, free cash flow for the third quarter of 2025 was positive $10.625 billion, and the cumulative free cash flow for the first nine months narrowed from $29.508 billion in the same period of 2025 to $11.049 billion.
This structural reversal in cash flow marks Meta’s official transition from a profit accumulation phase to a heavy capital investment phase, fundamentally changing its cash consumption model.
The expansion of the balance sheet further underscores the profound impact of infrastructure investments. As of the end of the third quarter, the company’s net property and equipment increased from $176.4 billion at the end of 2025 to $253.222 billion, reflecting large-scale construction of data centers, servers, and network infrastructure.
Meanwhile, long-term debt increased from $58.744 billion to $80.922 billion, indicating the company is leveraging debt to accelerate its computing power deployment. The negative free cash flow does not signify a liquidity crisis, as operating cash flow remains robust; however, this trend suggests that Meta is rapidly converting current advertising profits into future computing power assets. The key variables going forward will be whether these new assets can gradually generate revenue and profits, and whether operating cash flow can continue to grow to meet expanding capital needs.
Woofun AI’s compiled data shows that, despite mounting financial pressures, Meta’s management has not signaled any intention to scale back; instead, it has clearly outlined a strategy of continued aggressive investment through 2027. Chief Financial Officer Susan Li noted that while the 2027 budget is still being finalized, the company expects to maintain its aggressive investment pace. For the fourth quarter, the company forecasts revenue between $73 billion and $76 billion, with a midpoint guidance of $74.5 billion—a sequential increase of approximately 15.3% from the third quarter. With reallocated resources, losses at Reality Labs are expected to be contained, with full-year 2026 losses projected to decline to $18 billion–$19 billion, and further narrowing to around $15 billion in 2027. Regarding capital expenditures, Meta has raised the lower end of its 2026 full-year capex guidance by $5 billion, updating the range to $135 billion–$145 billion.
This upward adjustment reflects the company's strong confidence in the long-term value of AI infrastructure, prioritizing the maintenance of a leading position in computing power reserves even amid short-term cash flow pressures.
On AI product progress, the commercial testing of the Muse assistant and the release of the model matrix have become new growth engines. In just one and a half months since its launch in the U.S. and Canada, over 10 million users have set up Muse, with approximately 25% using it daily. The most common use case is shopping, accounting for over 25% of usage, with an average spending amount exceeding $200. Due to first-week transaction volumes far surpassing expectations, Meta had to partner with Stripe to temporarily increase credit limits to meet the surge in transaction demand. Moving forward, Muse will expand to Meta’s smart glasses starting next month, broadening its hardware access points. In model development, Meta released its first proprietary frontier model, Muse Wonder, which ranks among the most advanced in the industry across multiple key metrics.
At the same time, Muse Spark 1.2/1.3, Muse Code, and two open-weight open-source models—Muse Glimmer and Muse Spark Open—were also released.
The release of this product portfolio marks Meta's shift from purely infrastructure investment to commercial exploration at the application layer, aiming to generate new revenue streams through personal agents.
To support its long-term strategic planning, Meta has made significant adjustments to its governance structure by issuing non-voting Class C common stock to reserve capital capacity. The board has approved the creation of a new class of non-voting Class C common stock, to be distributed as a stock dividend to existing Class A and Class B shareholders at a rate of one Class C share for each share held.
This arrangement will not alter the existing voting rights structure of shareholders. The company’s special committee has signed an agreement with Zuckerberg aimed at maintaining the current governance structure. This move follows the example set in 2014 by Alphabet (GOOGL.US), Google’s parent company, which completed a stock split by distributing non-voting Class C shares proportionally, enabling founders Larry Page and Sergey Brin to retain control of the company while continuing to issue new shares for employee equity incentives and acquisitions. For Meta, the non-voting Class C shares will primarily be used for future employee equity incentives, potential strategic acquisitions, or other capital initiatives.
This optimized governance structure aims to expand financing tools and empower the company with the flexibility to make strategic investments driven by AI, while preserving the founder-led leadership structure and ensuring sufficient capital and governance space for substantial long-term investments.
The clarification of the technology roadmap reveals the bottlenecks in computing power, the pursuit of improved cost efficiency, and the long-term vision for agent-based commerce. Zuckerberg stated that the primary limitation of Muse currently is its high computational intensity, which is why it has not yet been widely launched outside the U.S. and Canada; the company is working to enhance computational efficiency and build sufficient computing capacity to support a user base of over one billion in the future.
Although Muse is currently a compute-intensive application, its cost efficiency is expected to improve by more than 10 times by 2027, enabling profitability at scale. On the enterprise front, Meta is collaborating with partners including Sierra, Walmart (WMT.US), Stripe, and Shopify (SHOP.US) to develop the Personal Agent Protocol, defining secure and reliable methods for consumer Muse agents to directly negotiate and transact with enterprise agents. In model development, Meta plans to distill state-of-the-art models into smaller, more efficient versions, aiming to deliver equivalent intelligence within a year at a fraction of the current cost of Muse Wonder.
This technical approach indicates that Meta is attempting to address the high costs of AI applications through model distillation and efficiency optimization, thereby building a sustainable business loop.
Meta’s long-term strategic bet is that the additional revenue and efficiency gains from AI will be sufficient to sustain this multi-year infrastructure expansion. In 2027, Zuckerberg will continue to pursue aggressive investment, relying on advertising revenue as the financial foundation while using AI infrastructure development to create new growth opportunities.
Although AI investments take longer to generate returns than core business operations, they have already driven significant incremental revenue for core advertising by improving recommendations and monetization efficiency. As investment continues to grow, the company must demonstrate that the additional revenue and efficiency gains from AI can offset rising infrastructure costs and achieve profitability over the coming years. This represents Meta’s next major strategic transformation following the mobile internet era, and its success will determine the company’s ultimate position in the AI era.