Meta is not making a speculative bet that AI might improve its business someday.
AI is already improving recommendations, engagement, and advertising performance.
The revenue proves it.
But the infrastructure bill is arriving just as quickly.
That is the whole debate.
At $747.82, investors are betting that Meta can keep compounding its core advertising franchise while spending more than $100 billion a year on the systems meant to extend it.
What Meta actually controls
Meta controls attention at global scale.
Facebook, Instagram, WhatsApp, and Messenger connect billions of people. Meta's recommendation engines decide what much of that audience sees. Its advertising system converts those signals into measurable customer acquisition.
In Q2 2026, revenue rose 28% to $60.8 billion. Ad impressions increased 14%, and average price per ad rose 12%.
Why does that matter?
Because Meta is getting both more inventory and more value from each unit of inventory.
Family daily active people reached 3.60 billion.
The network remains enormous, and the monetization engine is accelerating.
Business quality versus stock valuation
Meta's core business is one of the best cash-generating advertising systems ever built.
It owns global distribution, proprietary engagement data, auction infrastructure, and the recommendation models that increasingly determine user experience.
That quality is visible in 28% revenue growth.
It is less visible in current free cash flow.
Operating cash flow was $31.9 billion in Q2, but free cash flow was only $784 million after $30.1 billion of property purchases and $962 million of finance-lease principal payments.
Management now expects 2026 capital expenditures of $130 billion to $145 billion.
Why does that matter?
Because Meta is using almost all the cash from today's advertising machine to build tomorrow's AI machine.
The return on that transfer will determine the stock.
What $747.82 requires
I use 2031 as the checkpoint and a 10.1% required annual return.
That hurdle begins with the 5.11% 10-year Treasury yield, adds the 4.14% implied U.S. equity risk premium, and uses a 1.20 bottom-up beta for advertising and digital platforms.
At a 25 times terminal earnings multiple, the current price requires roughly $48.40 of 2031 EPS.
Meta earned $6.18 per diluted share in Q2, despite legal and severance charges.
The required outcome is plausible.
It is not conservative.
It asks Meta to absorb a historic investment cycle and still nearly double quarterly earnings power over five years.
Three possible outcomes
Conservative: about $346 today
Meta earns $28 per share in 2031 and receives a 20 times terminal multiple.
Advertising remains healthy, but depreciation and infrastructure costs rise faster than monetization. Reality Labs continues consuming cash, and the market assigns Meta a lower multiple.
The business remains profitable.
The capital cycle overwhelms the stock return.
Base case: about $649 today
Meta earns $42 per share in 2031 and receives a 25 times terminal multiple.
AI keeps improving engagement and ad conversion, capital spending moderates, and new products contribute without damaging core margins.
That is a strong operating outcome.
It remains below the checked price.
Upside case: about $1,004 today
Meta earns $58 per share in 2031 and receives a 28 times terminal multiple.
This requires AI to strengthen advertising, messaging to become meaningfully monetized, and today's infrastructure to support new revenue streams with attractive incremental margins.
At $747.82, the market is pricing an outcome between my base and upside cases.
Debt, dilution, and the infrastructure bill
Meta's Q2 Form 10-Q showed $90.3 billion of cash and marketable securities and $84.0 billion of long-term notes.
The company issued roughly $25 billion of debt in May and did not repurchase shares during the first half of 2026.
It also used $8.7 billion of cash for taxes related to employee stock awards.
Why does that matter?
Because the old Meta capital story was simple: generate enormous free cash flow and buy back stock.
The new story is more complicated: generate enormous operating cash flow, reinvest nearly all of it, add debt, and ask shareholders to wait for the return.
That can be rational.
It must also be measured.
What to watch out for
- Ad impressions and price per ad - because both volume and monetization must keep compounding.
- Capital expenditures and free cash flow - because infrastructure spending cannot outrun cash generation forever.
- Depreciation growth - because the income-statement cost of today's build arrives after the cash leaves.
- WhatsApp and business messaging monetization - because Meta needs additional returns from its installed attention base.
- Reality Labs operating losses - because long-duration optionality still has a real annual cost.
- Debt, repurchases, and diluted shares - because capital allocation has shifted from shrinking the share count to funding the build.
What could change the thesis
The thesis would strengthen if ad growth stays strong while capex peaks, free cash flow recovers, and messaging or AI services become visible profit contributors.
The thesis would weaken if revenue growth slows while capex and depreciation remain elevated, or if Meta repeatedly adds debt without a measurable rise in owner earnings.
The disconfirming evidence would be better AI products but worse cash economics per share.
Bottom line
Meta controls global attention and one of the world's most efficient systems for monetizing it.
Core business quality: exceptional.
AI execution: already visible in revenue.
Capital intensity: historic.
Valuation: above my base case.
At $747.82, Meta is priced as though the infrastructure bill will create another profit engine, not merely defend the current one.
The practical principle: when a cash machine reinvests nearly every dollar, the question is no longer whether it can spend. It is what each new dollar earns.
Sources
- Meta Q2 2026 earnings release
- Meta Q2 2026 Form 10-Q
- Meta investor relations financials
- Federal Reserve H.15 interest rates
- Damodaran U.S. industry betas
- Damodaran implied equity risk premium
Scenario values are analytical estimates, not company guidance or price targets. Educational only - not financial advice.