Amid an unprecedented AI expansion, Meta is spending greater than most. The company is constructing two huge data centers The reporting indicates this There shall be as much as $600 billion in spending on U.S. infrastructure over the following three years.
These numbers may not cause a stir in Silicon Valley, but they’re beginning to make Wall Street nervous.
The problem got here to a head this week Meta reported quarterly resultsThis showed that the corporate's operating costs increased by $7 billion year-on-year, and capital expenditures were nearly $20 billion. This was the results of intensive investment in AI talent and infrastructure, which has not yet generated significant revenue for the corporate. If analysts I pressed for more detailsMark Zuckerberg made it clear that the spending had just begun.
“The right thing is to attempt to speed up this to ensure that we now have the computing power that we’d like, each for AI research and for brand spanking new things that we're doing, and to attempt to bring our core computing power posture to a unique level,” Zuckerberg told analysts on the conference call. “Our view is that if we bring the brand new models that we're developing in MSL in there and get real frontier models with novel capabilities that don't exist elsewhere, then I believe that's just an enormous latent opportunity.”
If his goal was to reassure investors, it didn't work. At the tip of the conversation, the worth of the Meta share had plummeted. Two days later, the defeat has only deepened. Meta stock fell 12% by Friday's closing bell, representing greater than $200 billion in lost market capitalization.
It's dangerous to read an excessive amount of into stock prices, and from a purely financial perspective, Meta's quarterly earnings weren't all that bad. (A quarterly profit of $20 billion is nothing to complain about.) But this was the primary quarter through which Meta's aggressive AI spending on each talent and infrastructure had a visual impact on the corporate's bottom line. Even more worrying was that, except for many massive data centers and well-paid AI researchers, it was not clear what the cash was actually buying.
Analysts pressed Zuckerberg on why he was spending a lot on AI and after they might expect to see revenue from the increasing spending. But the decision got here at an odd point in Meta's planning: There was no clear budget for planned spending and no available product to anchor a sales forecast. As a result, Zuckerberg was left with only general claims in regards to the promise of AI.
Techcrunch event
San Francisco
|
thirteenth–fifteenth October 2026
“There's going to be all styles of recent products around different content formats, and we're beginning to see that,” he said through the call. “And then there are the business versions of it, like Business A… the opposite part is that smarter models just improve the core business and improve the recommendations that we make across the family of apps and improve the recommendations in promoting.”
Meta isn't the one company spending billions of dollars on AI infrastructure. So it's price determining why the identical expenses don't spook investors at Google or Nvidia, each of which had an excellent quarter. OpenAI is the most important offender, spending the identical amount with far less financial cushion than Meta.
There is real concern that we’re making a bubble, and if that’s the case, Meta will weather things higher than most due to its core business.
But if you happen to ask Sam Altman why he spends lots of of billions of dollars on computers, he'll inform you that he runs one in every of the fastest-growing consumer services in human history — and one which brings in $20 billion in annual revenue. We can argue about how sustainable the expansion rate is (that's a separate blog post), but behind all of the OpenAI hype is definitely a fast-growing product. A rapidly growing ARR number is answer to all questions.
Meta doesn’t have such a product and it is just not clear where it should come from.
The company's strongest AI product is the Meta AI Assistant, which Zuckerberg said within the conference call has a couple of billion energetic users. But those numbers are actually boosted by the three billion energetic users on Facebook and Instagram, and it's hard to see the present version of Meta AI as a competitor to ChatGPT. There's also the Vibes video generator, which actually increased each day energetic users, but has limited business impact beyond that.
The most ambitious project is the Vanguard smart glasses Released earlier this month. However, the glasses feel more like an extension of Meta's Reality Labs work than a real try to harness the ability of LLMs.
To put it simply, these are promising experiments, not mature products.
It's telling that when pressed about infrastructure spending, Zuckerberg didn't point to recent launches but as a substitute focused on the following generation.
Zuckerberg, while emphasizing the approaching impact of the Superintelligence Lab's recent models, said he could be very enthusiastic about recent products.
“It’s not nearly Meta AI as an assistant,” he said. “We expect to construct novel models and novel products, and I sit up for sharing more when we now have it.”
But it was an earnings release and never a product launch, so all he could say was that there can be more to report “in the approaching months.”
As the market response showed, this answer is tenuous.
To be fair, it's only been 4 months since Zuckerberg restructured his company's AI team, and the brand new superintelligence team hasn't yet had time to bring a breakthrough AI product to market. But as the corporate spends billions of dollars to stay competitive in AI, there remains to be no clear indication of what role Zuckerberg desires to play in the brand new industry.
Will Meta AI leverage the corporate's granular personal data store to evolve right into a ChatGPT competitor? Is Vibes step one in a consumer entertainment game built on Meta's targeted promoting system? Or are Zuckerberg’s references to “business AI” perhaps hints at a more detailed corporate strategy?
So far it's anyone's guess. Whatever the reply, Meta is under pressure to seek out it – and shortly.
