How much is Meta spending on AI this year?

Meta expects to spend $130 billion to $145 billion on capital expenditures for the full year, mainly related to artificial intelligence, with the majority going toward frontier research and inference capacity for AI models.

Why is Meta considering selling compute capacity to other companies?

Meta's AI models do not see much external use, so the company is already building up excess capacity in data centers that is not being utilized. Zuckerberg said other AI companies have reached out to Meta to buy access to its compute capacity at a premium over the purchase price.

How is Meta's core advertising business performing?

Meta's revenue grew 28% year over year to $61 billion last quarter, mainly driven by advertising growth across Facebook, Instagram, and WhatsApp, driven by improvements in targeting technologies.

AI Stocks & MarketsAI Business & IndustryYahoo Finance AIAug 16, 2026Aug 17, 2026, 04:01 JST

Meta considers selling AI compute power as $130B–$145B capex plan strains earnings

Send on LINE

3 Key Points

What happened
Meta spent $30 billion on capital expenditures last quarter and plans to spend $130 billion to $145 billion for the full year, mostly on AI infrastructure. Zuckerberg's team is now exploring selling excess compute capacity to other AI companies, similar to deals SpaceX has signed that generate tens of billions in annual revenue.
Why it matters
Meta's heavy AI spending is outpacing its advertising revenue growth—operating margin fell from 43% a year ago to 31% in Q2 2026, and EBIT fell for the first time in a long while last quarter. The fact that Meta is considering selling compute power signals it cannot find internal uses for the AI infrastructure it has already invested tens of billions developing, a bearish indicator that depreciation will continue to pressure earnings ahead.
What to watch
Meta's core advertising business remains strong, with revenue growing 28% year over year to $61 billion last quarter, driven by targeting improvements across Facebook, Instagram, and WhatsApp. However, unless Meta can monetize its AI investments to generate tens of billions in revenue or reverse its aggressive capex plan, earnings are likely to continue shrinking, making the stock's $600 price tag less attractive despite its 22 price-to-earnings ratio relative to other tech peers.

Not sure about something? Ask the AI

Questions and answers are published on this page.

Summaries like this, in your inbox every morning.

Get news like this every morning, free

Context & Analysis

Meta's AI spending trajectory reveals a disconnect between ambition and return. The company deployed $30 billion in capex last quarter and plans to commit $130 billion to $145 billion annually, yet its AI chatbots command only a sliver of the market against Alphabet's Gemini, OpenAI's ChatGPT, and Anthropic's Claude. The shift toward selling excess compute capacity—compared to deals SpaceX has signed generating tens of billions annually—suggests Meta views this infrastructure not as a path to AI product dominance but as a cost-recovery mechanism. The financial strain is already visible: operating margin contracted from 43% a year ago to 31% in Q2 2026, and EBIT fell for the first time in a long while despite a 28% year-over-year revenue jump to $61 billion driven by advertising. The core advertising business remains robust and growing, but it cannot keep pace with the depreciation flowing through the income statement as capex compounds without offsetting AI revenue. Unless Meta either reverses these aggressive investments or monetizes them to generate tens of billions in new revenue, earnings are positioned to shrink over the next few years, making the stock's valuation relative to peers less compelling despite its sub-$600 price and 22 P/E ratio.

FAQ

How much is Meta spending on AI this year?: Meta expects to spend $130 billion to $145 billion on capital expenditures for the full year, mainly related to artificial intelligence, with the majority going toward frontier research and inference capacity for AI models.
Why is Meta considering selling compute capacity to other companies?: Meta's AI models do not see much external use, so the company is already building up excess capacity in data centers that is not being utilized. Zuckerberg said other AI companies have reached out to Meta to buy access to its compute capacity at a premium over the purchase price.
How is Meta's core advertising business performing?: Meta's revenue grew 28% year over year to $61 billion last quarter, mainly driven by advertising growth across Facebook, Instagram, and WhatsApp, driven by improvements in targeting technologies.

Yahoo Finance AIRead Original Article

Send on LINE

AI news that matters for your work, delivered every morning.

Pick your industry and the AI tools you use, and get news related to your work every day.

Get it free

Free · 30 seconds with Google · unsubscribe anytimeWhat is AIToday? →

Ask AI

Ask AI anything about this article. The AI reads this article, earlier AIToday articles, and Wikipedia, and cites its sources. Q&As are published on this page for other readers too.