Meta's AI Pivot: Turning Compute into a New Revenue Engine
Amid heavy capex and a stock drawdown, Meta is repositioning from an ads-only model to an AI infrastructure and enterprise agent platform—a strategic shift that could redefine its business mix.
META · Earnings Call · 2026-07-29
From Social Ads to AI Infrastructure
Meta’s Q2 2026 earnings call marked a clear inflection point. The company is no longer just a social media and advertising giant; it is aggressively repositioning as a foundational AI infrastructure player and enterprise agent platform. As Mark Zuckerberg framed it, the opportunity is now “massive” and spans “personal agents,” “business agents,” and a large enterprise opportunity—including selling compute directly. This shift is visible in the company’s own keyword trajectory: “enterprise” spiked to a momentum of 276 in Q2 2026, its highest level in over a year, while agent work and personal agent consistently rank among the top company-specific themes.
Susan Li highlighted the tangible early results: “For the first time, quarterly Family of Apps other revenue reached $1 billion and grew 73% year-over-year, driven primarily by WhatsApp paid messaging and subscriptions revenue.” That mark, while small relative to the $60.4 billion in total Family of Apps revenue, signals a newly material non-advertising revenue stream. More consequential, however, is Meta’s stated intent to treat its compute capacity as a monetizable asset, with Zuckerberg noting, “a significant portion of our compute is going to go towards training our models… but we also expect to grow a large business serving large customers as well.” Sources of capital have become a strategic lever; the company announced a strategic venture with BlackRock to develop a 1-gigawatt data center in El Paso, Texas.
The Compute Monetization Engine
Meta is bridging its traditional ad business with a new compute-for-hire model. The company is already receiving “a lot of offers for compute at a significant premium over what we paid for it,” as Zuckerberg admitted. Yet management is careful to frame this as a long-term option rather than a near-term windfall. The key distinction is between selling compute directly versus selling intelligence (models, agents, and APIs). Zuckerberg expects the latter to have “significantly higher margin,” so the near-term focus is on scaling products that deliver ROI—like business agents, which now have more than 1 million businesses using them weekly, and the recently launched Meta Business Agent platform.
Susan Li elaborated on the capital strategy:
We've also been evolving our capital structure in recent years to include a greater mix of debt as we work to bring down our cost of capital, and we have generally found it prudent to continue adding cost-efficient long-duration sources of capital as we make investments in initiatives that themselves have long time horizons, especially AI infrastructure projects.
This marks a notable departure from previous quarters. In the Q1 2026 call, Li had already indicated that the company “continued to underestimate our compute needs” and was exploring external partnerships. Now those explorations have crystallized into concrete structures. The BlackRock deal and similar partnerships are designed to attract outside capital, reducing the burden on Meta’s own free cash flow while still allowing Meta to control the buildout. This is a direct answer to investor concerns about the sustainability of the $130–145 billion 2026 capex plan.
The Cost of Ambition
The strategic pivot does not come cheap. While Q2 revenue grew 28% year-over-year to $60.8 billion, operating income fell 8% on a GAAP basis, driven by a 55% jump in total expenses. Legal charges and severance expenses compounded the pressure. The company’s fundamental trajectory mirrors this tension: operating margin dropped 2.8 percentage points year-over-year to 40.6% in Q2 2026, and free cash flow margin fell below 13%—its lowest level in years.
Meta’s operating margin declined from 43.1% in Q1 2025 to 40.6% in Q2 2026, even as revenue growth accelerated. The gap between topline growth and profitability underscores the heavy investment cycle.
The market has taken notice. META's stock is down 12.7% over the past 90 days, with a drawdown of over 20% from its April high. The fundamental data shows a clear pattern: capital expenditure jumped 47% year-over-year to $19.0 billion in the latest quarter, while free cash flow (less SBC) fell to $7.2 billion, only slightly above the year-ago level. Investors are wrestling with a new tradeoff: Meta is buying growth through compute, but the returns on that compute remain uncertain.
Zuckerberg defended the approach, citing the long-term optionality:
“There's just nowhere near enough compute for all the demand. So that is why we see that like basically, we are getting a large number of offers for the compute that we have, but also we have a lot of internal uses that we think are going to be quite valuable.” — Mark Zuckerberg, Chief Executive Officer (CEO) · 2026-07-29
He also argued that the company is uniquely positioned to build the full stack—from silicon to models to consumer products—a theme echoed in prior quarters. In the Q1 2026 call, Susan Li had said, “Our expectation is that compute will become even more central to the business going forward,” and the current results validate that vision.
Community and Product Momentum
Amid the capex story, the underlying business remains robust. Daily active users reached 3.6 billion, Instagram surpassed 2 billion daily actives, and Threads crossed 500 million monthly actives. Ad revenue still grew 27%, and advertiser performance improved, with Meta Generative Recommender and Muse Image adoption scaling. The company also launched Meta One, a subscription tier, and expanded WhatsApp business agents globally.
The user-level metrics are encouraging: “9 million small businesses on our platforms are now using at least one of our AI ad creative tools,” noted Zuckerberg. This adoption is feeding the flywheel of better recommendations and higher ad ROAS, which supports the core business while the enterprise bets mature.
Yet the pivot also carries execution risk. The agentic capabilities are still nascent, and the enterprise compute market is crowded. Meta’s competitive advantage will hinge on its distribution reach—3.6 billion users and millions of businesses—and its ability to integrate AI seamlessly into the most-used apps. As Zuckerberg put it, “We're the only major company building AI with the primary goal of putting superintelligence directly into people's hands.”
Conclusion
What changed at Meta is not just a new product line; it’s a fundamental reallocation of capital and ambition. The company is deliberately trading near-term margin for long-term optionality in AI infrastructure and agentic services. The fundamental numbers confirm the shift: total revenue is scaling, but so is capex and the drag on free cash flow. The market’s negative reaction reflects the uncertainty inherent in that trade. However, if Meta can convert its compute into a new revenue stream at even a fraction of its ad business efficiency, the payoff could be substantial. The next few quarters will be critical in showing whether the enterprise and compute monetization can turn from promise into profit.