Meta just bumped its 2026 capex forecast up to as much as $145 billion for the AI boom—and investors flinched

Meta Platforms is splashing some serious cash on AI infrastructure, and investors have flinched. 

The company reported first quarter 2026 earnings results on Wednesday and raised its full-year 2026 capital expenditure guidance to $125 billion to $145 billion, up from a previous range of $115 billion to $135 billion. Meta told investors the boost was the result of higher prices for components and “additional data center costs to support future-year capacity.” 

Last year, Meta spent $72.2 billion on capex, up roughly $30 billion from the year before. The company is now guiding to nearly double what it spent in 2025, and more than it spent in 2025 and 2024 combined.

In after-hours trading, the stock tumbled more than 6% as a result of the jump in capex guidance. In contrast, Alphabet and Amazon—which are also spending enormous sums on AI infrastructure buildout, and which both announced earnings on Wednesday—saw their share prices rise after hours, in part because they both reported AI-related growth in their massive cloud-services businesses.

Asked about ROI, Zuckerberg says it’s ‘a very technical question’

Zuckerberg pointed to “memory pricing” as a driver of the higher costs and he attempted to soothe investors by explaining how he expects the spending plan to pan out.  

“Every sign that we’re seeing in our own work and across the industry gives us confidence in this investment,” said Zuckerberg. “That said, we are very focused on increasing the efficiency of our investments, and as part of that, we are rolling out more than one gigawatt of our own custom silicon that we’re developing with Broadcom as well as significant amount of AMD chips to complement the new Nvidia systems we’re rolling out as well.”

Zuckerberg was asked during the call to explain any signposts or key factors he is watching to ensure Meta is “on the right path” to generating a healthy return on the investment over the next 12 to 24 months in Meta AI, new advancements or to its core algorithm.

“That’s a very technical question,” Zuckerberg responded. “The things that we’re watching are to make sure that we’re on track to building leading models and leading products. The formula for our company has always been to build experiences that can get to billions of people and focus on monetizing them once you get to scale.”

He added that he doesn’t think Meta has “a very precise plan for exactly how each product is going to scale month over month, or anything like that, but I think we have a sense of the shape of where these things need to be.”

“I’m quite comfortable that the lab we’re building is on track to be a leading lab in the world,” said Zuckerberg. 

Revenue and profit climb sharply

Meta reported Q1 revenues of $56.3 billion, up 33% from the same period a year ago. Operating income rose 30% to $22.9 billion, and profits grew 61% to $26.8 billion. The company noted that profits got a boost from an $8 billion tax benefit in the first quarter, which helped offset a $15.9 billion tax charge in the third quarter of 2025 when the One Big Beautiful Bill Act took effect. 

Total expenses in the first quarter ballooned 35% to $33.4 billion, driven mostly by infrastructure costs and employee compensation, said chief financial officer Susan Li. Meta doled out a series of stock option grants to Li and other executives targeting a $9.46 trillion market capitalization, a feat no company has ever achieved. 

“The growth in infrastructure costs was due to higher depreciation data center operating costs and third-party cloud spend,” said Li. “The growth in employee compensation was driven by technical hires we’ve added over the past year, particularly AI talent.”

Li also noted the company shared internally that it would “reduce the size” of Meta’s employee base in May. The company reportedly plans to slash hundreds of jobs in the U.S. and abroad among teams including sales, recruiting, and on its hardware unit. 

Meta, like other major tech firms, has been pouring money into data centers and servers to train its AI models, which it views as essential to its core advertising business and longer-term investments in personal AI agents for business, health, and entrepreneurship. Zuckerberg has said the investments will strengthen the ad business by making recommendations more relevant and improving the way ads are targeted to increase the time consumers spend on its platforms including Instagram, WhatsApp, and Facebook. 

On the earnings call, Zuckerberg said its new AI models will help the company evolve beyond looking at statistical patterns showing the types of people engaging with content.

“For the first time in Meta’s history, we’re going to be able to develop a first-principles understanding of what you care about and what each piece of content in our system is about,” he said. “So that way, we can show you more useful things for what you’re trying to accomplish and we’ll also be able to create personalized content specifically for people to help you achieve your goals as well.”

Melissa Otto, head of Visible Alpha Research at S&P Global, said the downturn in the stock price after hours was a clear reaction to the increase in capex guidance. It was already “pretty high” said Otto, and the company had a good quarter, “but it wasn’t a blowout.”

“It raises this question about what is the real ROI on all this capex that they’re spending,” said Otto. “I think the investment community is getting a little frustrated at the amount of cash they’re burning.”

Otto said investors are on the lookout for information about how Meta’s investment in AI infrastructure is contributing to top-line and efficiency gains. 

During his remarks, Zuckerberg said the Superintelligence AI lab released “significantly upgraded” version of Meta AI, which was its first. 

“Over the past 10 months, we have built the strongest research team in the industry and established the scientific and technical foundations to scale very advanced models,” said Zuckerberg. “Now that we have a strong model, we can develop more novel products as well.”

This story was originally featured on Fortune.com