AI: Meta & 'Zuck' aiming for the AI Clouds. AI-RTZ #1135
Took a couple of years, but my then suggestion of Meta proactively building an AWS type of cloud business off its massive AI Data Center Infrastructure investments, is closer to reality.
Meta is likely to follow Elon Musk’s SpaceX/xAI in renting its excess AI cloud compute for top dollar in the supply constrainted environment this AI Tech Wave.
Bloomberg highlights the key points in “Meta Is Planning a Cloud Business to Sell AI Computing Power”:
“Meta Platforms Inc. is developing plans for a cloud infrastructure business that will sell access to AI computing power and models, setting up a new vector of competition with industry leaders like Amazon Web Services, Microsoft Azure and Google Cloud.”
“Meta, which has been rushing to secure expensive data centers and other infrastructure to fuel its own artificial intelligence ambitions, is forming a business to generate revenue from excess computing power sold to outside customers, according to people familiar with the matter, who asked not to be named as the details aren’t public.”
Then a discussion of the next logical step. Meta, with Elon Musk’s SpaceX/xAI/Tesla recently, would join Amazon AWS, Microsoft Azure and Google Cloud to the AI Cloud business this AI Tech Wave. And of course the ‘Neocloud’ companies like CoreWeave, Nebius and others. In the all important Box no 2 of the AI Tech Stack.
“One potential plan includes selling access to various AI models that are hosted on Meta’s existing AI infrastructure, an approach similar to AWS’s Bedrock offering, the people said. Meta would run the data centers and chips that power the models, including its own Muse Spark models, and charge developers to access them.”
“The company is also considering selling access to “raw” computing capacity, akin to other so-called neocloud businesses like CoreWeave Inc., the people said. Development of these new business lines is part of Meta Compute, an internal initiative to build and manage the company’s AI infrastructure efforts, according to a person familiar with the plans. Meta Compute is led by Santosh Janardhan, Meta’s head of infrastructure; Daniel Gross, a leader inside the Meta Superintelligence Labs AI unit; and Meta President Dina Powell McCormick.”
Investors like what they see here for now, in Mark Zuckerberg’s long AI game.
“Shares of Meta jumped 9.3% to $615.55 at 10:04 a.m. in New York on Wednesday, the biggest intraday gain since April. CoreWeave fell as much as 14%. Nebius Group NV, a Dutch AI data center company that trades in New York, fell as much as 17%.”
The strategy builds of course on top of Meta’s aspirations around its Meta Superintelligence (MSL) business I’ve writtena about at length.
“Meta has made developing AI “superintelligence” a top priority, and has committed hundreds of billions of dollars to data centers and other AI infrastructure, like expensive chips that it deems necessary to make that happen. That investment, which has left investors anxious about Meta’s plans to earn a return on that spending, includes major computing deals with CoreWeave, Alphabet Inc.’s Google and Oracle Corp., among others.”
“A cloud business offers one way to return some of that investment. AWS, Azure and Google Cloud have spent decades building platforms that rent access to computing power, storage and software over the internet — businesses that now command tens of billions of dollars per quarter in revenue.”
It’s all driven off one general assumption that is becoming the ‘crowded trade’ of the day. Especially with Meta’s Manhattan sized multi-gigawatt AI Data Centers to come.
“The demand for computing power from major AI developers remains insatiable. Meta and other tech companies have committed tens of billions of dollars for data center capacity for their own needs in recent quarters, adding to the massive sums the industry is spending on artificial intelligence. Selling computing is just another means of capitalizing on the broader AI boom.”
“As appetite for AI has surged, those providers have also expanded to rent the specialized chips and computing capacity needed to train and run AI models. It is a complex business, requiring not only vast fleets of data centers but also software platforms, enterprise sales teams and customer support operations.”
And as mentioned above, Elon Musk already has shown how this could be done adroitly in today’s market. As I discussed not too long ago.
SpaceX/xAI days before its mega-AI IPO signed a total of $70+ billion multi-month contracts with Anthropic ($40+ billion) and Google ($30+ billion). Both deals are cancelable by either side at short notice. Note that Google is also an early shareholder of SpaceX, and has a board seat on the company. So some ‘circular’ characteristics to that mutually opportunistic deal. As Bloomberg lays it out,
“Elon Musk’s SpaceX, which acquired his AI startup xAI in February, recently emerged as a key player in this space, renting access to its massive data center in Memphis to Anthropic PBC earlier this year and striking a deal with Google. That strategy could help xAI generate more than $50 billion in revenue by 2028 and $100 billion by 2030, according to an estimate by Bloomberg Intelligence.”
“Despite the complexities, Meta Chief Executive Officer Mark Zuckerberg has signaled to investors that he’s open to selling excess computing infrastructure, or even a so-called API service where customers would pay for AI usage — a business that’s usually measured in “tokens,” or the amount of data used and generated for a customer query.”
And Mark himself underlines the obvious. Leveraging the huge number of Nvidia infrastructure he has bought to date, from his good buddy Jensen Huang.
““It’s definitely on the table,” Zuckerberg said during a call with shareholders in May. “Almost every week there are different companies that come to us from the outside asking us to both stand up an API service or asking if we have compute that they could buy from us at some premium to what we’ve bought it at.”
“We haven’t done that yet because we think we have a use for the compute,” Zuckerberg said at the time. “But obviously if we get to a point where we feel that we have overbuilt, then that is an option that we have, and that is partially what gives us confidence in investing in building this out.”
“Amid a fast-moving AI race, Zuckerberg has repeatedly suggested that he believes the industry is constrained when it comes to computing capacity and that Meta should amass as much as possible and determine its use later.”
Coming back to Meta, even though it does not have an IPO ahead of it, founder/CEO Mark Zuckerberg is planning raising tens of billions in equity and/or debt like Google, Amazon and other big tech peers for continued, ramping investments in AI cloud data center infrastructure.
Back in 2024, I said of Meta’s AWS style opportunity, just as Amazon itself was figuring out its AI opportunities back then.
“Meta has many possible AI futures to be shaped.”
“Meta with its current AI assets and execution disposition, may be in the right place at the right time. Much like Amazon was with AWS in 2004. The opportunity to scale its LLM AI software and compute infrastructure assets into an entirely new business is palpably similar. “
“ Mark Zuckerberg, is uniquely empowered to make these opportunities happen. “
And that founder/CEO is now ready to make his AWS style moves. I’ll call them ZWS for ‘Zuck Web Services, just like I’ve been calling SpaceX’s AWS style efforts ‘EWS’, for ‘Elon Web Services’.
For both founder/CEOs, signaling plans to earn extra revenues Amazon AWS style from its existing capacity, makes tactical and strategic sense. And they’re both now leaning into the AI cloud strategy this AI Tech Wave. Stay tuned.
(NOTE: The discussions here are for information purposes only, and not meant as investment advice at any time. Thanks for joining us here)