During the announcement of its Q2 2026 financial results Meta CEO Mark Zuckerberg revealed that the company is transforming itself from a Social Media Company into an AI Company, aiming to develop technology spanning from foundational infrastructure to AI directly experienced by users. However, the market has not immediately embraced this strategic bet.
Following the earnings announcement, Meta's shares plunged sharply in after-hours trading, dropping as much as 11%, due to investor disappointment with both profit figures and revenue outlooks. The bigger question waswhere Meta is deploying its massive funds and when AI will begin generating returns?
The most striking figure in this report is not revenue but free cash flow remaining after operating expenses, which stood at only $784 million, a steep decline from $8.55 billion in the same quarter last year—the lowest in at least five years according to the company’s financial data.
This quarter, the company reported total revenue of $60.8 billion, a 28% increase year-over-year, with approximately 98% of revenue still coming from advertising on Facebook and Instagram. Meanwhile, net profit fell 14% to $15.8 billion, with earnings per share at $6.18, below market expectations of $7.22.
The Reality Labs unit, responsible for VR, AR, Mixed Reality, Quest, and wearable devices including Smart Glasses and related AI technologies, remains the heaviest operating loss maker, generating only $431 million in revenue and incurring a loss of $4.6 billion.
Meanwhile, total costs and expenses in Q2 surged 55% to $42.03 billion. Although some of this increase stemmed from special charges—such as $2.4 billion in legal expenses and $1.18 billion in severance costs—the free cash flow figure of $784 million has become a key focus for investors.
For this quarter, Meta announced an increase in its full-year capital expenditure budget to $130–145 billion, up from the previous range of $125 billion set just three months ago, stating that most of this will be invested in AI infrastructure to meet rising AI demand.
The company is advancing multiple Data Center projects, including the 1GW Meta Compute Data Center under a strategic partnership with BlackRock valued at $14 billion in El Paso, Texas; the Hyperion project in Louisiana costing over $50 billion; and a $9 billion Data Center in Alberta, Canada.
This clearly shows that Meta's AI capital expenditures are not merely for purchasing GPUs or training models but involve massive investments in physical infrastructure to support long-term AI development, reflecting Mark Zuckerberg's new AI strategy vision.
Zuckerberg stated that AI is already integrated into the existing business, embedding large language models (LLMs) into Instagram and Facebook recommendation systems to better understand content, interests, and user goals, enabling the system to present more relevant content.
In advertising, Meta uses LLMs to predict and rank ads shown to users by incorporating behavioral context from both general content and advertisements, making ads more targeted and increasing conversions. He emphasized that AI investments are beginning to yield returns, with Meta’s advertising revenue growing faster than that of other companies reporting in the sector.
Although nearly all revenue still comes from advertising, Meta is striving to become an AI company structurally and investment-wise by building a new engine.
This includes foundational infrastructure such as Data Centers, Compute, self-designed chips, AI models, software, and AI Agents as new company products, aiming to build an AI infrastructure capable of supporting billions of users and expanding into new future business areas.
Zuckerberg also revealed that one goal is to offer Meta’s AI models to other companies, especially enterprise customers. Recently, they launched Muse Spark 1.1, described by Alexandr Wang, Meta’s Head of AI, as the company’s most capable model for Agentic AI and coding tasks, priced cheaper than models from OpenAI and Anthropic.
Muse Spark 1.1 will be available via Public API, with plans to expand access through partnerships and coding agents, alongside developing tools to make it easier for companies to adopt the models. This reflects Meta’s effort to convert AI investments into direct revenue.
He also discussed Personal AI Agents as the next objective, noting that Meta is the only major tech company developing AI with the primary goal of putting superintelligence hardware directly into people’s hands.
He envisions that in the near future, users will have AI Agents working on their behalf 24/7, with the challenge being to make these AI tools accessible and easy to use for billions of people without requiring technical knowledge as coding agents do today. Meta aims to make this a new product stage, forming the foundation and next revenue stream for the company.
Another noteworthy development is Meta’s effort to convert AI infrastructure costs into revenue-generating assets. Unlike Alphabet, Amazon, and Microsoft, which have large cloud businesses supporting AI investments, Meta lacks a comparably large cloud business.
Zuckerberg said Meta is receiving offers from companies wanting to rent Meta’s compute infrastructure at prices significantly above Meta’s own costs, potentially becoming a new way to monetize its massive investments.
Meta’s compute resources are used for AI model training, core business expansion, personal agent development, and new products, but the company also expects to build a large enterprise business by providing services to clients. Beyond APIs and business agents, Meta plans to develop additional coding and productivity tools.
If successful, Meta will not only be a company that buys compute to develop AI but one that can sell surplus compute capacity to others, generating returns from its invested infrastructure.
Analysts have compared the current situation to Meta’s prior Metaverse investments, where the company spent tens of billions of dollars creating virtual worlds and VR devices before the market adopted them broadly, resulting in underwhelming outcomes.
The key difference now is that AI is already generating revenue for many companies and can immediately enhance Meta’s core advertising business. However, investors still want to see how these investments will translate into revenue, profit, and cash flow.
Meta is at a critical crossroads, determining whether AI will become a new engine of growth or another high-stakes gamble by Mark Zuckerberg, similar to the Metaverse.
Read more
Source information Meta , CNBC , Yahoo Finance , Bloomberg
Follow the Facebook page: Thairath Money at this link -