The $135 Billion Bet: What Meta's Q3 Earnings Reveal About the AI Spending Reckoning

There is a number that Wall Street is sitting with this Thursday morning: 135. That is how many billion dollars - at the low end - Meta Platforms now expects to spend on capital expenditures in 2026, after the company raised the floor of its guidance range by $5 billion in its third-quarter earnings report on Wednesday evening. The update arrived alongside a revenue beat, a blowout Q4 outlook, and a $9.89 billion legal settlement charge that obscured what was, by almost any measure, a strong quarter. The market's reaction will be the first real test of whether investors have finally made peace with Meta's AI spending machine - or whether the capex number is still the one that moves the stock.

The Numbers That Matter

Meta reported Q3 2026 revenue of $64.6 billion, up 26.1 percent year over year and ahead of the $63.35 billion analyst consensus. Adjusted earnings per share came in at $7.40, beating the $6.54 estimate by $0.86. GAAP earnings per share were $3.83, well below expectations, but the gap is entirely explained by a one-time $9.89 billion charge tied to a settlement with 52 state attorneys general over teen mental health and social media litigation - a charge that had been widely telegraphed and that Bank of America had already modeled into its reduced EPS forecast. Excluding the charge, the quarter was a clean beat on both the top and bottom lines.

The Q4 guidance was the number that will likely define the stock's reaction. Meta guided for fourth-quarter revenue of $73 billion to $76 billion, a range that implies roughly 22 to 27 percent year-over-year growth and sits above the $72 billion consensus. For a company that has now posted four consecutive quarters of revenue growth above 20 percent, the guidance is a statement of confidence in the advertising business that funds everything else Meta is building.

The Capex Question That Never Goes Away

The more complicated story is the one that has been following Meta for the better part of two years: what does it cost to build the AI infrastructure behind the advertising machine, and when does that spending start generating returns that show up in free cash flow?

Meta narrowed its 2026 capital expenditure guidance to $135 billion to $145 billion, raising the floor from $130 billion. The company has spent approximately $85.8 billion through the first three quarters of the year, which means the fourth quarter alone will require somewhere between $49 billion and $59 billion in additional spending to hit the range. That is not a rounding error. It is a single-quarter capital deployment that would exceed the annual capex of most Fortune 500 companies.

The free cash flow math is stark. In Q3, Meta generated $32.8 billion in operating cash flow and spent $34.9 billion on capital expenditures - meaning its AI buildout consumed more cash than its business produced in the quarter. That dynamic, which first appeared in Q2 when free cash flow collapsed to $784 million despite $60.8 billion in revenue, is the central tension in the Meta investment thesis. The advertising business is exceptional. The infrastructure bill is enormous. The question is whether the returns on that infrastructure will arrive before the market loses patience with the spending.

Muse and the Monetization Question

Mark Zuckerberg used the earnings call to make the case that the returns are already beginning to materialize. The centerpiece of that argument is Muse, Meta's AI personal assistant that launched in September and reached 10.9 million downloads in its first 40 days - a milestone that Sensor Tower noted was reached by only 11 other apps in history. Muse has surpassed ChatGPT's initial mobile adoption rate and is now the 21st most-advertised brand in the United States by impression volume, up from 73rd in September.

Zuckerberg described the quarter as delivering "major milestones toward personal superintelligence for everyone," citing the launch of Muse and Muse Wonder, Meta's first frontier-class AI model. The framing is deliberate: Meta wants investors to see the company not as an advertising business that is burning cash on AI experiments, but as an AI company that happens to generate extraordinary advertising revenue while it builds toward something larger.

The monetization path for Muse is still being constructed. Meta's "other revenue" category - which includes WhatsApp paid messaging, subscriptions, and early AI product revenue - reached $1 billion in Q2, up 73 percent year over year. That is a real number, but it is a rounding error relative to the $130 billion-plus in advertising revenue the company is on track to generate in 2026. The gap between what Meta is spending to build AI and what it is currently earning from AI is the defining financial question of the next two years.

What the Legal Settlement Tells You

The $9.89 billion charge deserves more attention than it has received in the earnings coverage. The settlement with 52 attorneys general resolves litigation related to teen mental health and social media - a legal and regulatory overhang that has been building since the Senate hearings of 2023 and 2024. The total settlement value is up to $18.59 billion, paid over ten years, with the Q3 charge representing the present value of that obligation.

The settlement is significant not just for its size but for what it represents: a resolution of the most politically charged regulatory risk Meta has faced in years. With 52 state AGs signing on, the deal effectively closes the book on the most aggressive domestic legal threat to Meta's core business model. That is worth something to investors who have been discounting Meta's valuation for regulatory risk - and it is worth noting that Meta chose to take the charge in a quarter where the underlying business was strong enough to absorb it without derailing the narrative.

The Broader Earnings Season Context

Meta's results arrive in the same week as Tesla's Q3 report, which showed revenue of $28.05 billion in line with estimates, adjusted EPS of $0.48 beating the $0.43 consensus, and gross margins recovering to 19.6 percent from 16.8 percent in Q2. Tesla's robotaxi miles grew eightfold quarter over quarter, providing the first concrete evidence that its autonomous driving ambitions are generating real operational scale. Both reports, taken together, suggest that the Magnificent Seven earnings season is off to a stronger start than the market's recent selloff implied.

The more important context is the FOMC meeting on October 27 and 28, where the Federal Reserve will decide whether to raise rates again in a market where the 10-year Treasury yield is sitting above 5 percent and core inflation remains above 3 percent. Meta's Q4 guidance of $73 billion to $76 billion is a statement that the advertising market is not breaking under the weight of elevated rates - that consumer spending, which funds the ad budgets of Meta's customers, is holding up well enough to support continued double-digit revenue growth. That is a meaningful data point for the Fed's assessment of economic resilience, and it is a meaningful data point for investors trying to decide whether the AI spending cycle has legs into 2027.

What Investors Should Watch

The specific signals to watch in the coming days are clear. First, whether Meta's stock reaction on Thursday reflects a market that has finally accepted the capex trajectory as the cost of building a durable AI platform, or whether the $49 billion to $59 billion Q4 spending implied by the guidance triggers another round of free cash flow anxiety. Second, whether Zuckerberg provides any early signal on 2027 capex - the number that analysts currently model at $193 billion to $225 billion - in post-earnings commentary or investor conferences. Third, whether the Muse adoption metrics translate into any measurable revenue contribution in Q4 guidance, or whether monetization remains a 2027 story.

The $135 billion bet is not a gamble in the conventional sense. Meta's advertising business generates enough cash to fund the buildout - just barely, and with less margin for error than the company had two years ago. What it is, is a statement of conviction: that the companies willing to spend at this scale on AI infrastructure will emerge with competitive advantages that cannot be replicated by those who waited. Whether that conviction is rewarded depends on whether Muse and the AI products that follow it can close the gap between what Meta is spending and what it is earning. The Q3 report suggests the advertising engine is strong enough to keep the bet alive. The Q4 report, and the 2027 capex guidance that will accompany it, will tell investors whether the returns are finally arriving.