Elon Musk recently announced what seemed like a strange pairing of his companies: SpaceX was acquiring xAI, the artificial intelligence company that also owns the X social media platform.
The merger combines a highly profitable rocket company with an AI startup that is spending billions of dollars as it competes with OpenAI and other rivals for artificial intelligence dominance.
The merger also comes as SpaceX – whose main business recently has been launching Musk’s Starlink broadband satellites – prepares for an IPOmaybe for this year.
The combined companies are valued at $1.25 trillion. That’s higher than the recent separate valuations of SpaceX at $800 billion and xAI’s holding company at $230 billion. The IPO shares are expected to cost around $525.
“I think this provides a narrative for the merger of SpaceX and xAI, a sort of cash-generating company and a company that needs liquidity – at a pretty spectacularly high valuation,” said Steve Collar, president of Swissto12, a Swiss satellite maker.
While the merger may make financial sense, it goes well beyond that. Here’s what you need to know about the plan.
This seems like an unlikely marriage. Why is Musk doing this?
Besides the financial advantages, there are technical reasons. Musk outlined a futuristic plan that involves sending up to a million satellites – called data centers – into an orbit where the sun could power them day and night as they perform artificial intelligence calculations.
What exactly is a satellite data center?
The basic concept is simple. The satellite’s main function would be to perform calculations on specialized chips, powered by solar panels in a “sun-synchronous orbit” that would keep them exposed 24/7 to the sun’s powerful rays.
Aren’t hundreds of billions of dollars already being spent building AI data centers on Earth?
Musk insists that within three years it will be cheaper to have AI data centers in space than the hyperscale data centers currently being built on Earth due to strain on the power grid – a timeline that some experts have questioned.
But don’t get started 1 millions of satellites in orbit, is that crazy?
Some think so. Indeed, the number of proposed launches is nearly 100 times greater than the approximately 11,000 Starlink broadband satellites that SpaceX has sent into space since 2019 on its Falcon 9 workhorse.
However, Musk plans to use his enormous Space rocketwhich is under development and can lift more than five times the payload.
“If you look at deploying a whole bunch of additional satellites, you create a huge need for Starship,” Collar said.
Still, this seems like an excessive number of launches.
There is no doubt about it. In a blog postMusk has talked about hourly launches to put millions of tons of satellites into orbit every year.
However, Collar said a working network of AI data centers in space could be much smaller, with 10,000, 50,000 or as many as 100,000 satellites. “I think with Elon, it’s a little bit of this all-or-nothing thing,” he said.
Where is the development of space AI data centers?
Little. It was only in November that Google announced Suncatcher Projectwhich it says is exploring an “interconnected network of solar-powered satellites” equipped with its own chips. And Starcloud, a one year old startup in Redmond, Washington, backed by Andreessen Horowitz and other major venture capital firms, launched a demonstration satellite in November.
“I would say you’re talking about the 2040 timeline before you have a usable system,” Collar said.
What are the technological challenges?
The first hurdle to overcome is developing a computer chip that works in the extreme conditions of space, which includes exposure to much higher levels of radiation.
“If you want to take an Nvidia chip and make it work in space, you’ll probably have to invest a few years and several million dollars to make it so-called space qualified to ensure it’s not damaged by radiation,” said Ehud Behar, an astrophysicist at the Technion – Israel Institute of Technology.
Are there any downsides to a sun-synchronous orbit?
Yes. This exposes satellites to a lot of heat and the way to dissipate the heat generated by the sun and computer chips would be to use large aluminum panels called radiators, but they are inefficient, Behar said.
“You’re going to have to build large areas to dissipate the heat. That’s part of the challenge. I don’t think it can be these small satellites. The mass has to be large enough to make it cost effective,” he said.
How does SpaceX propose to bring all this data back to Earth?
In a request from the Federal Communications CommissionSpaceX describes how the centers would transmit data via optical or laser signals to the Starlink network for transmission to the ground while the entire system is in low orbit about 300 miles to 1,200 miles above Earth.
Is there a technological challenge to this as well?
Yes. Collar said satellites in such an orbit circle the globe about every 90 minutes, making it difficult for them to communicate with each other. A better idea would be to place the ground communications satellite in a higher orbit, where it would circle the Earth at a much slower rate and its transmissions to the ground would cover a larger area.
“But launching into a lower Earth orbit costs a lot less, so it’s all a gigantic (compromise) and that’s what makes it really interesting,” he said.
In a blog post, Musk also explained how everything these the data will eventually be accessible on mobile phones from anywhere on Earth over a network it is building. How would this work?
SpaceX purchased $17 billion in wireless spectrum in September of EchoStar, thus strengthening a partnership already established with T-Mobile which allowed customers to send a text message when they are off-grid. This followed with an additional $2.6 billion purchase of EchoStar spectrum a month later.
However, Craig Moffett, telecommunications analyst at MoffettNathansonsaid that Elon Musk’s satellite network, given the distances involved, would be far less efficient than ground-based cellular networks and would more benefit regions without access to existing networks.
Starlink doesn’t have “the right set of assets to serve more than a niche player in the wireless market,” he said.