Meta's Stock Tumbles as AI Spending Crushes Free Cash Flow

Meta’s stock took a beating in extended trading Wednesday, plunging nearly 10% after the social media giant delivered disappointing quarterly results and signaled slower revenue growth ahead. The company’s aggressive pivot toward artificial intelligence and massive data center buildouts is clearly taking a toll on its balance sheet, raising questions about whether this spending spree will ultimately pay off.

The damage came on multiple fronts. Meta guided revenue for the current quarter at $61 billion to $64 billion, with a midpoint of $62.5 billion. That’s below analyst expectations of $63.15 billion. Meanwhile, daily active people across Meta’s family of apps hit 3.6 billion, slightly missing Wall Street’s estimate of 3.61 billion.

But the real alarm bell? Free cash flow collapsed to just $784 million in the second quarter, down from $8.55 billion a year earlier. That’s a stunning 91% decline, largely driven by Meta’s relentless investment in AI infrastructure and computing capacity.

The Capital Expenditure Spiral

Meta narrowed its full-year capital expenditure guidance to a range of $130 billion to $145 billion, up significantly from the prior guidance of $125 billion to $145 billion. To put that in perspective, the company is essentially doubling down on its most bullish spending scenario. CEO Mark Zuckerberg defended the strategy on the earnings call, explaining that a significant portion of compute will go toward training models, growing the core business, and developing personal AI agents.

The company isn’t shy about its ambitions in this space. Meta recently hired Alexandr Wang, the founder and CEO of Scale AI, in a deal involving a $14.3 billion investment. Wang’s track record in AI infrastructure development suggests Meta is deadly serious about competing with OpenAI, Anthropic, and other AI leaders.

Just this month, Meta released the Muse Spark 1.1 model, which Wang described as the “strongest model for agentic and coding work yet” at a lower price point than competitors. That’s the carrot dangling in front of investors: maybe all this spending will eventually unlock a lucrative new revenue stream.

Monetizing Compute Capacity

Here’s an intriguing development: Zuckerberg revealed that Meta is fielding serious inquiries about leasing out excess computing capacity at a premium. “We’re getting a lot of offers for compute at a significant premium over what we paid for it,” he said. That suggests Meta could potentially offset some of its massive capital outlays by selling capacity to other organizations.

The company recently announced a $14 billion data center partnership with BlackRock in El Paso, Texas, following earlier disclosures of a $50 billion Hyperion project in rural Louisiana and a $9 billion facility in Alberta, Canada. These aren’t incremental investments, they’re structural bets on becoming a dominant force in AI infrastructure.

Metric-wise, second-quarter results showed total costs and expenses surged 55% year over year to $42.03 billion. Stripping out $2.4 billion in legal charges and $1.18 billion in severance costs from the May layoffs, operating income actually grew 9% year over year, according to CFO Susan Li. Net income still declined to $15.85 billion from $18.34 billion, or $7.14 per share.

The Reality Labs Wildcard

One relative bright spot: Meta’s Reality Labs unit, which develops VR and AI-powered wearables, posted a $4.6 billion operating loss while generating $431 million in revenue. That actually beat analyst expectations for a $5.07 billion loss on $423.4 million in revenue. Progress, though it’s still a massive money pit.

Meta shares are down 11% this year, underperforming the Nasdaq by roughly 16 percentage points. Investors are clearly wrestling with whether this AI bet justifies the near-term pain. Rival Alphabet just reported negative free cash flow for the first time due to AI spending, so Meta isn’t alone in this predicament. The critical difference: Alphabet, Amazon, and Microsoft all have thriving cloud businesses to offset infrastructure costs. Meta doesn’t, which makes its compute-leasing ambitions all the more critical to watch.

The question hanging over technology investors now is whether Meta can actually monetize this infrastructure advantage before the market loses patience with the burn rate.

Source: CNBC