Loading video...

Video Failed to Load

There was a problem loading this video. This could be due to a temporary network issue or the video might be unavailable.

Data centers on the ground are limited by permits and power. SpaceX's answer is Starmind, orbital AI data centers with about 72 Nvidia chips and up to 175 kW per satellite. That's about 5,700 satellites per gigawatt. Terafab builds the chips, and Starship launches them. Jo Bhakdi thinks it... show more

Herbert Ong

106,566 subscribers

488,463 views • 7 days ago •via X (Twitter)

News & Politics Science & Technology

Anya Rossi• Live Now

Private livecam show

32 Comments

Elon Musk6 days ago

True

Simon Manning7 days ago

You need ground based production to put the intelligence into effect. That is required to PAY for the intelligence and power put in space. So ground based production MUST keep up with space based intelligence. That is not a trivial point. Tesla cant use all of the intelligence in space.

Martians 🪐6 days ago

@elonmusk Mars Data Center @elonmusk

ANA COUPER ♃ 𒀭 𒀉 𒀭𒊹6 days ago

@elonmusk Reminds me of the paper I wrote for this in 2023 for @SpaceX

jay🥷6 days ago

i have been researching and studying a bunch since starship's successful orbit and deployment... it's insane what people are doing with their lives right now while this is literally right around the corner

ᒍOᕼᑎ ᗪEᒪᗩᑎO6 days ago

How do you cool them?

Âxel ⟢6 days ago

@elonmusk Why Elon is tweeting no?

Iscanic6 days ago

And that, Son, is how Skynet was created...

Kevin Brown6 days ago

Absurdly too ambitious sorry. Scaling on that level is mid-2030's at the earliest.

The Office of Fury 🔥6 days ago

I just want a StarCell (phone) with StarSearch engine so the government can’t control me. The UK needs it!

Αλέξιος6 days ago

Clown. How you will cool that data centers in space?

Julian C6 days ago

Permits on Earth? Just go to space, I guess.

JoAnne Songs6 days ago

@elonmusk Unlimited space for data centers 👍

Annie▫️6 days ago

@elonmusk Come teach us some of these stuffs in Nigeria @elonmusk

STOO6 days ago

@elonmusk Data centers for Iowa

Kelly Talbert6 days ago

@elonmusk Ok.. and what is the ultimate purpose? I'm only on board if it's for improvement and not destruction of anything. I mean it. For memory, for quantum computing .. and intelligence agrees

Becker Meister7 days ago

5700 sats per gigawatt and I still can't get a rack in Frankfurt on time

Blackwell7 days ago

Your posts constantly surface stocks and ideas I wouldn’t have found on my own, and @JavierMolinaSTK is another analyst I learn from every week. You two are my favorite follows for stocks.

M.Dong.6 days ago

听着就令人激动

William Johnson7 days ago

It is that simple....and that hard! But Muskonomies thrive on doing hard stuff.....making it look easy!

Su-Anne de Jong6 days ago

Awesome!! 😎 Super excited about the SI Data Centres 🚀🤩 @elonmusk Please recruite me☺️

BraveTom6 days ago

Good move

ikeoscar6 days ago

@elonmusk Good one

George Holdcroft6 days ago

How do orbital data centers expel heat? That seems like the most difficult challenge. Maybe there is a way of recycling it.

Gracodex6 days ago

How can you cool down the data center because that's the most important thing.

Swik7 days ago

👍

Still The Original Mack6 days ago

I was thinking of locating them on the Moon

Ruby boy6 days ago

Hmmm

Linda6 days ago

@elonmusk 🚀🚀🚀🚀

John Utah6 days ago

@elonmusk Terrestrial power and permits are a bottleneck. Moving compute to orbit completely changes the constraints

Marina Lehner6 days ago

Starmind - that's the way. Orbital SI data center by @SpaceX ftw! 🚀🦾✨️

Kristian Blåsol6 days ago

And that way, there will be no kill switch when the AI goes rogue. Unlimited solar power with no way to turn it off except shooting them down, which will be interpreted as a hostile move... The all seeing eye will hover in the sky above us and control us all 😅🤪

Related Videos

NEWS: SpaceX has detailed Starmind, its plan for up to 1 million AI compute satellites in orbit, on a new website. Starmind moves AI data centers into space. Each satellite runs AI compute in orbit and beams results to Earth through Starlink lasers. The satellites draw constant solar power and cool for free in the vacuum. SpaceX filed that plan with the FCC in January. The first satellite, AI1, carries 150 kW of peak compute and spans 70 meters. Starship will carry 30 to 50 AI1 satellites per launch. SpaceX will build them at a new Gigasat factory in Bastrop, Texas. The AI chips come from a planned SpaceX and Tesla fab called Terafab. Two prototypes target early 2027, with thousands of satellites from late 2027.

NEWS: SpaceX has detailed Starmind, its plan for up to 1 million AI compute satellites in orbit, on a new website. Starmind moves AI data centers into space. Each satellite runs AI compute in orbit and beams results to Earth through Starlink lasers. The satellites draw constant solar power and cool for free in the vacuum. SpaceX filed that plan with the FCC in January. The first satellite, AI1, carries 150 kW of peak compute and spans 70 meters. Starship will carry 30 to 50 AI1 satellites per launch. SpaceX will build them at a new Gigasat factory in Bastrop, Texas. The AI chips come from a planned SpaceX and Tesla fab called Terafab. Two prototypes target early 2027, with thousands of satellites from late 2027.

Muskonomy

26,421 views • 3 months ago

Here’s Elon Musk explaining that SpaceX could launch up to one terawatt per year of AI compute capacity into Earth orbit — equivalent to the power output of roughly 1,000 nuclear power plants. “The next step beyond Earth data centers is our Earth orbital data centers. And we'll be launching with SpaceX orbital data centers at the 100-to-200-gigawatt-per-year level. Not cumulative, per year. And ultimately, we see a path to maybe launching as much as a terawatt per year of compute from Earth. But what if you want to go beyond a mere terawatt per year? In order to do that, you have to go to the Moon. So by having factories on the Moon, building AI satellites, and having a mass driver, which is the kind of thing you really only learn about in or read about in science fiction, but we're going to make it real. We're actually going to have a mass driver on the Moon. And if you do that, you can go several orders of magnitude greater. You can go to 1,000 gigawatts or more per year and ultimately get to maybe a millionth and then a thousandth and maybe even a few percent of the Sun's energy.” xAI All-Hands, February 10, 2026

Here’s Elon Musk explaining that SpaceX could launch up to one terawatt per year of AI compute capacity into Earth orbit — equivalent to the power output of roughly 1,000 nuclear power plants. “The next step beyond Earth data centers is our Earth orbital data centers. And we'll be launching with SpaceX orbital data centers at the 100-to-200-gigawatt-per-year level. Not cumulative, per year. And ultimately, we see a path to maybe launching as much as a terawatt per year of compute from Earth. But what if you want to go beyond a mere terawatt per year? In order to do that, you have to go to the Moon. So by having factories on the Moon, building AI satellites, and having a mass driver, which is the kind of thing you really only learn about in or read about in science fiction, but we're going to make it real. We're actually going to have a mass driver on the Moon. And if you do that, you can go several orders of magnitude greater. You can go to 1,000 gigawatts or more per year and ultimately get to maybe a millionth and then a thousandth and maybe even a few percent of the Sun's energy.” xAI All-Hands, February 10, 2026

ELON CLIPS

21,962 views • 6 months ago

Elon Musk and Jensen Huang did the math on AI power at the White House, and every number pointed the same direction: The US grid runs on about 500 gigawatts. Musk: every 5 GW added is +1% power use, and he'd "bet anyone" it maps to roughly +1% GDP. 10 GW is +2%. Jensen's number: about $60B of economic output per gigawatt, per year. 10 GW from SpaceX alone would be about 25% of all the power added in America. China produces 3x the electricity of the US. Musk thinks the US figures out chips. Generation is the constraint. Jensen: 10 to 20 GW a year of buildout, about 1 million jobs. "We used to call them data centers. These are really SI factories." First time in 50 years the US is re-industrializing. Where the next wave comes from: SpaceX and $TSLA target 200 GW of solar per year, launched to orbit by Starship, where panels deliver nameplate or better. "The math maths."

Elon Musk and Jensen Huang did the math on AI power at the White House, and every number pointed the same direction: The US grid runs on about 500 gigawatts. Musk: every 5 GW added is +1% power use, and he'd "bet anyone" it maps to roughly +1% GDP. 10 GW is +2%. Jensen's number: about $60B of economic output per gigawatt, per year. 10 GW from SpaceX alone would be about 25% of all the power added in America. China produces 3x the electricity of the US. Musk thinks the US figures out chips. Generation is the constraint. Jensen: 10 to 20 GW a year of buildout, about 1 million jobs. "We used to call them data centers. These are really SI factories." First time in 50 years the US is re-industrializing. Where the next wave comes from: SpaceX and $TSLA target 200 GW of solar per year, launched to orbit by Starship, where panels deliver nameplate or better. "The math maths."

WOLF

60,893 views • 8 days ago

Elon Musk: “The next step beyond Earth data centers is Earth orbital data centers. We’ll be launching SpaceX orbital data centers at the 100 to 200 gigawatt per year level — not cumulative, per year. Ultimately we see a path to maybe launching as much as a terawatt per year of compute from Earth. But if you want to go beyond a terawatt per year, you have to go to the moon. By having factories on the moon building AI satellites, and having a mass driver — something you usually only read about in science fiction but we’re going to make it real. If you do that, you can go several orders of magnitude greater, to 1,000 gigawatts or more per year, and ultimately maybe even a few percent of the sun’s energy. It’s difficult to imagine what an intelligence of that scale would think about, but it’s going to be incredibly exciting.”

Elon Musk: “The next step beyond Earth data centers is Earth orbital data centers. We’ll be launching SpaceX orbital data centers at the 100 to 200 gigawatt per year level — not cumulative, per year. Ultimately we see a path to maybe launching as much as a terawatt per year of compute from Earth. But if you want to go beyond a terawatt per year, you have to go to the moon. By having factories on the moon building AI satellites, and having a mass driver — something you usually only read about in science fiction but we’re going to make it real. If you do that, you can go several orders of magnitude greater, to 1,000 gigawatts or more per year, and ultimately maybe even a few percent of the sun’s energy. It’s difficult to imagine what an intelligence of that scale would think about, but it’s going to be incredibly exciting.”

Mars University

39,804 views • 8 days ago

NEWS: SpaceX just made its AI space computer 67% more powerful, Elon Musk says The satellite is called AI1. It is basically a data center that floats in space and runs on sunlight instead of the power grid. Musk says its peak power is going up to ~250 kW, from 150 kW. Its average power is going up to 160 kW, from 120 kW. Why the boost. A single Nvidia Rubin NVL72 rack, 72 of Nvidia's newest AI chips, needs that much power to run. The bigger number means one AI1 can now handle a full rack. Musk's reasoning is simple. A GPU running all day uses about two-thirds of its peak power, even when it is very efficient. So the satellite has to be built for the peak. AI1 works with chips from any maker, not just Nvidia. SpaceX has filed to launch up to 1 million of these satellites. (Source: Elon Musk on X, July 16)

NEWS: SpaceX just made its AI space computer 67% more powerful, Elon Musk says The satellite is called AI1. It is basically a data center that floats in space and runs on sunlight instead of the power grid. Musk says its peak power is going up to ~250 kW, from 150 kW. Its average power is going up to 160 kW, from 120 kW. Why the boost. A single Nvidia Rubin NVL72 rack, 72 of Nvidia's newest AI chips, needs that much power to run. The bigger number means one AI1 can now handle a full rack. Musk's reasoning is simple. A GPU running all day uses about two-thirds of its peak power, even when it is very efficient. So the satellite has to be built for the peak. AI1 works with chips from any maker, not just Nvidia. SpaceX has filed to launch up to 1 million of these satellites. (Source: Elon Musk on X, July 16)

Muskonomy

56,507 views • 2 months ago

Elon Musk says Orbital AI Data Centers will be easier than Communication Satellites for SpaceX: "The Starlink V3 communications satellites is an incredibly complex machine. The AI Data Center will be much simpler by comparison. It’s really just solar power, plus radiator, some basic equipment, and the laser links would connect to the Starlink communications constellation, and then to the ground."

Elon Musk says Orbital AI Data Centers will be easier than Communication Satellites for SpaceX: "The Starlink V3 communications satellites is an incredibly complex machine. The AI Data Center will be much simpler by comparison. It’s really just solar power, plus radiator, some basic equipment, and the laser links would connect to the Starlink communications constellation, and then to the ground."

Nic Cruz Patane

1,256,820 views • 4 months ago

Terafab is starting to sound like a lot more than just a chip factory! Tesla and SpaceX are reportedly putting $16.8B into phase one, with plans to handle everything from chip design and fabrication to memory, packaging and testing. The idea is basically to bring pieces of what TSMC, Micron and NVIDIA do under one roof. And the chips could eventually power everything from Tesla cars and Optimus to SpaceX data centers and AI compute in space. $TSLA $SPCX

Terafab is starting to sound like a lot more than just a chip factory! Tesla and SpaceX are reportedly putting $16.8B into phase one, with plans to handle everything from chip design and fabrication to memory, packaging and testing. The idea is basically to bring pieces of what TSMC, Micron and NVIDIA do under one roof. And the chips could eventually power everything from Tesla cars and Optimus to SpaceX data centers and AI compute in space. $TSLA $SPCX

Herbert Ong

31,110 views • 1 month ago

🇺🇸 Wall Street put $1/2T dollars behind the AI power bill. NVIDIA locked in Apollo, BlackRock, Blackstone, Brookfield, Goldman, and KKR to raise more than $500B for data centers and chips. Larry Fink’s numbers: the U.S. needs over 70 gigawatts, each one running $50-60B. He is selling the concrete, the steel, the electricians and the 3M worker hours that go into every 100 megawatts. The negative headlines about data centers miss the industrial build that is actually happening. Writer: Lucas

🇺🇸 Wall Street put $1/2T dollars behind the AI power bill. NVIDIA locked in Apollo, BlackRock, Blackstone, Brookfield, Goldman, and KKR to raise more than $500B for data centers and chips. Larry Fink’s numbers: the U.S. needs over 70 gigawatts, each one running $50-60B. He is selling the concrete, the steel, the electricians and the 3M worker hours that go into every 100 megawatts. The negative headlines about data centers miss the industrial build that is actually happening. Writer: Lucas

Mario Nawfal

40,998 views • 1 month ago

John Fetterman claims backlash to data centers is "driven by the CPP" and says Trump is right about them

John Fetterman claims backlash to data centers is "driven by the CPP" and says Trump is right about them

Aaron Rupar

206,917 views • 1 month ago

ELON MUSK ANNOUNCED A CHIP FACTORY 10 TIMES THE SIZE OF TESLA'S GIGAFACTORY 🤯 HE CALLS IT THE TERAFAB. THE ENTIRE GLOBAL AI CHIP INDUSTRY IS ON TRACK TO PRODUCE AROUND 100 GIGAWATTS OF COMPUTE PER YEAR. MUSK WANTS A SINGLE FACTORY TO PRODUCE 1 TERAWATT PER YEAR. THAT'S 10X THE OUTPUT OF THE ENTIRE GLOBAL INDUSTRY. THE GOAL? BUILD ENOUGH AI COMPUTE TO POWER ORBITAL DATA CENTERS. MOST PEOPLE THINK IT'S A DECADE AWAY. MUSK THINKS IT STARTS NEXT YEAR. THIS IS ABSOLUTELY CRAZY.

ELON MUSK ANNOUNCED A CHIP FACTORY 10 TIMES THE SIZE OF TESLA'S GIGAFACTORY 🤯 HE CALLS IT THE TERAFAB. THE ENTIRE GLOBAL AI CHIP INDUSTRY IS ON TRACK TO PRODUCE AROUND 100 GIGAWATTS OF COMPUTE PER YEAR. MUSK WANTS A SINGLE FACTORY TO PRODUCE 1 TERAWATT PER YEAR. THAT'S 10X THE OUTPUT OF THE ENTIRE GLOBAL INDUSTRY. THE GOAL? BUILD ENOUGH AI COMPUTE TO POWER ORBITAL DATA CENTERS. MOST PEOPLE THINK IT'S A DECADE AWAY. MUSK THINKS IT STARTS NEXT YEAR. THIS IS ABSOLUTELY CRAZY.

Vivek Sen

84,878 views • 2 months ago

Dr. Eric Schmidt says AI needs “energy in all forms: renewable, non-renewable, whatever.” “It needs to be there, and it needs to be quickly.” “People are planning 10 gigawatt data centers.” “An average nuclear power plant in the United States is one gigawatt.” “Data centers will require an additional 29 gigawatts of power by 2027 and 67 more gigawatts by 2030.” “These things are industrial at a scale I have never seen in my life.” This is exactly why I believe the market is still underestimating the AI power and infrastructure opportunity in companies like $TE, $BE, $PUMP, $PSIX, $SMR, $FCEL, $MARA, $KEEL, and $OKLO. The biggest constraint to scaling AI remains electricity. The companies building power generation, grid infrastructure, electrical equipment, cooling systems, and AI data centers could be among the biggest beneficiaries over the next decade. Not financial advice. Do your own research.

Dr. Eric Schmidt says AI needs “energy in all forms: renewable, non-renewable, whatever.” “It needs to be there, and it needs to be quickly.” “People are planning 10 gigawatt data centers.” “An average nuclear power plant in the United States is one gigawatt.” “Data centers will require an additional 29 gigawatts of power by 2027 and 67 more gigawatts by 2030.” “These things are industrial at a scale I have never seen in my life.” This is exactly why I believe the market is still underestimating the AI power and infrastructure opportunity in companies like $TE, $BE, $PUMP, $PSIX, $SMR, $FCEL, $MARA, $KEEL, and $OKLO. The biggest constraint to scaling AI remains electricity. The companies building power generation, grid infrastructure, electrical equipment, cooling systems, and AI data centers could be among the biggest beneficiaries over the next decade. Not financial advice. Do your own research.

Leopold Stock Tracker

20,643 views • 3 months ago

Elon: To understand the universe, we have to explore it. Earth is a dust mote. The Sun holds 99.8% of the solar system’s mass. Orbital data centers are next. 100–200 gigawatts per year. Moon factories. Mass drivers. AI satellites into deep space. This is how consciousness scales. From dust mote to stars. Elon Musk

Elon: To understand the universe, we have to explore it. Earth is a dust mote. The Sun holds 99.8% of the solar system’s mass. Orbital data centers are next. 100–200 gigawatts per year. Moon factories. Mass drivers. AI satellites into deep space. This is how consciousness scales. From dust mote to stars. Elon Musk

Mario Nawfal

125,184 views • 6 months ago

There’s plenty to debate on data centers… But the data moratoriums taking over the country are driven by politics, not facts. The most advanced data centers use zero water and continue to become more energy efficient. And when data centers are banned… communities walk away from lower taxes, funded schools, and loads of good paying jobs. That's the trade off…

There’s plenty to debate on data centers… But the data moratoriums taking over the country are driven by politics, not facts. The most advanced data centers use zero water and continue to become more energy efficient. And when data centers are banned… communities walk away from lower taxes, funded schools, and loads of good paying jobs. That's the trade off…

Tim Young

11,081 views • 2 months ago

Elon Musk announced a chip factory 10 times the size of Tesla's Gigafactory. The goal is to produce enough AI compute to equal twice the entire electricity consumption of the United States. He called it the Terafab. Here is the number that stopped me cold. The entire global AI chip industry right now is on track to hit around 100 gigawatts per year of compute. Every Nvidia GPU, every Google TPU, every chip from every company on earth combined. 100 gigawatts. Musk wants one factory to produce a terawatt per year. A terawatt is 1,000 gigawatts. Ten times the output of the entire global industry. From a single building. To put the scale in physical terms, the Terafab would need to be around 100 million square feet. You would need Starship point to point transport just to get from one end to the other. But the reason for the scale is not ambition for its own sake. To launch meaningful AI compute into space, you need a billion chips per year running at a kilowatt each. That is not a number the current industry can produce. The Terafab is the only way to get there. The timeline he put out: a gigawatt of space AI compute annualized by end of next year. Then 10x per year from there. 10 gigawatts by year two. 100 gigawatts by year three. A terawatt beyond that. Most people think orbital data centers are a decade away. Musk is building the factory to make them possible by next year.

Elon Musk announced a chip factory 10 times the size of Tesla's Gigafactory. The goal is to produce enough AI compute to equal twice the entire electricity consumption of the United States. He called it the Terafab. Here is the number that stopped me cold. The entire global AI chip industry right now is on track to hit around 100 gigawatts per year of compute. Every Nvidia GPU, every Google TPU, every chip from every company on earth combined. 100 gigawatts. Musk wants one factory to produce a terawatt per year. A terawatt is 1,000 gigawatts. Ten times the output of the entire global industry. From a single building. To put the scale in physical terms, the Terafab would need to be around 100 million square feet. You would need Starship point to point transport just to get from one end to the other. But the reason for the scale is not ambition for its own sake. To launch meaningful AI compute into space, you need a billion chips per year running at a kilowatt each. That is not a number the current industry can produce. The Terafab is the only way to get there. The timeline he put out: a gigawatt of space AI compute annualized by end of next year. Then 10x per year from there. 10 gigawatts by year two. 100 gigawatts by year three. A terawatt beyond that. Most people think orbital data centers are a decade away. Musk is building the factory to make them possible by next year.

Ihtesham Ali

318,691 views • 3 months ago

In AI, the bottleneck isn't talent or models anymore. It's compute. Jo Bhakdi points to a claim that Anthropic and OpenAI could slow down sharply if they can't secure enough compute next year. Elon's reply is that SpaceX becoming the most valuable AI company is "one of the possible outcomes" if it reaches 10 gigawatts. On the ground, a gigawatt of AI compute costs $40 to $50 billion today. So 10 GW would be $400 to $500 billion of data centers. Elon says orbital compute gets much cheaper by 2028. The bigger play is SpaceX selling that compute, on the ground and in orbit, instead of keeping it all. Demand isn't slowing either. Nvidia is reportedly planning a big testing framework for AI agents, and Anthropic's IPO filing spends a lot of pages on agent risk. Testing all of that takes even more compute. $SPCX $TSLA

In AI, the bottleneck isn't talent or models anymore. It's compute. Jo Bhakdi points to a claim that Anthropic and OpenAI could slow down sharply if they can't secure enough compute next year. Elon's reply is that SpaceX becoming the most valuable AI company is "one of the possible outcomes" if it reaches 10 gigawatts. On the ground, a gigawatt of AI compute costs $40 to $50 billion today. So 10 GW would be $400 to $500 billion of data centers. Elon says orbital compute gets much cheaper by 2028. The bigger play is SpaceX selling that compute, on the ground and in orbit, instead of keeping it all. Demand isn't slowing either. Nvidia is reportedly planning a big testing framework for AI agents, and Anthropic's IPO filing spends a lot of pages on agent risk. Testing all of that takes even more compute. $SPCX $TSLA

Herbert Ong

11,462 views • 7 days ago

Oregon-based startup Panthalassa is developing floating AI data centers powered by wave energy. The 85-meter-long system uses seawater to cool AI chips while turbines generate electricity from ocean waves. A prototype called Ocean-2 is already being tested off the coast of Washington, with AI data transmitted through SpaceX's Starlink network.

Oregon-based startup Panthalassa is developing floating AI data centers powered by wave energy. The 85-meter-long system uses seawater to cool AI chips while turbines generate electricity from ocean waves. A prototype called Ocean-2 is already being tested off the coast of Washington, with AI data transmitted through SpaceX's Starlink network.

Space and Technology

553,419 views • 2 months ago

$SPCX Starmind AI satellites could route data through Starlink’s laser mesh avoiding complex onboard antennas and most direct spectrum requirements. That lets Starmind piggyback on more than 10,000 Starlink satellites with V3 models potentially carrying 2–3 Tbps each. This creates a distribution moat no standalone orbital-compute player can match but it also shifts the bottleneck to the ground by funneling enormous traffic through SpaceX’s gateways. Which is why SpaceX is expanding into E-, V- and W-band backhaul with D-band proposed for AI while new FCC rules replacing 1990s power limits could unlock up to 7x more capacity from the same satellite network.

$SPCX Starmind AI satellites could route data through Starlink’s laser mesh avoiding complex onboard antennas and most direct spectrum requirements. That lets Starmind piggyback on more than 10,000 Starlink satellites with V3 models potentially carrying 2–3 Tbps each. This creates a distribution moat no standalone orbital-compute player can match but it also shifts the bottleneck to the ground by funneling enormous traffic through SpaceX’s gateways. Which is why SpaceX is expanding into E-, V- and W-band backhaul with D-band proposed for AI while new FCC rules replacing 1990s power limits could unlock up to 7x more capacity from the same satellite network.

Shay Boloor

69,920 views • 2 months ago

ARK: SPACEX IS NO LONGER A ROCKET COMPANY After the largest IPO in history, the pitch to investors was about AI, not rockets: - SpaceX now tells investors more than 90% of its future market is AI, not space, after acquiring Elon Musk's xAI, its Grok models, and two of the most powerful data centers on Earth - The strategy is vertical integration: own the launch, the data centers, and the AI, with compute already rented to customers like Anthropic, Google, and Reflection AI - ARK estimates that at scale, sub $100 per kilogram launch costs make orbital data centers 25% cheaper to build than on the ground, in constant sunlight with nearly free energy

ARK: SPACEX IS NO LONGER A ROCKET COMPANY After the largest IPO in history, the pitch to investors was about AI, not rockets: - SpaceX now tells investors more than 90% of its future market is AI, not space, after acquiring Elon Musk's xAI, its Grok models, and two of the most powerful data centers on Earth - The strategy is vertical integration: own the launch, the data centers, and the AI, with compute already rented to customers like Anthropic, Google, and Reflection AI - ARK estimates that at scale, sub $100 per kilogram launch costs make orbital data centers 25% cheaper to build than on the ground, in constant sunlight with nearly free energy

Ark Invest Tracker

38,824 views • 2 months ago