Oncast Tech · Machines

SpaceX's AI now bigger than rockets and Starlink combined

· Fact-checked before it aired. Unconfirmed claims from this segment are not shown.

SpaceX's Super Intelligence segment grew revenue 830% year-over-year to $7.56 billion in Q3 2026, slightly more than the space launch ($1.77 billion) and connectivity/Starlink ($5.25 billion) segments combined ($7.02 billion).

Fact-check

VERIFIED TRUE

SpaceX's Super Intelligence segment grew revenue 830% year-over-year to $7.56 billion in Q3 2026, slightly more than the space launch ($1.77 billion) and connectivity/Starlink ($5.25 billion) segments combined ($7.02 billion).

SpaceX ☑️ The artificial intelligence division grew by 830% over the past year; it is now larger than the Starlink and rocket launch businesses combined. ☑️ Revenue by business segment for the last quarter: • Super Intelligence: $7.56 billion • Starlink: $5.25 billion • https://t

Original post · Richard Banks (@banks_abanks)

[cnbc.com](https://www.cnbc.com/2026/10/20/spacex-spcx-earnings-live-updates-q3-2026html.html)

Sources

What the hosts said

MAnd… just… an 830 percent jump in one year. Richard Banks posted a breakdown of SpaceX's business segments, and the AI division didn't just grow—it took over the whole company.
CWait, what does 'larger than Starlink and launch combined' actually mean? Because Starlink alone was the golden goose.
MHere's the numbers from last quarter: Super Intelligence pulled in seven and a half billion dollars. Starlink did five point two five billion. And the launch business came in at one point seven seven.
CSo the AI segment beat Starlink by over two billion, and if you add launch to Starlink, they still fall short of AI by a couple hundred million. That's just… staggering.
MThink about what that means for the company's identity. A rocket company whose biggest division is now artificial intelligence. It's like Boeing suddenly saying its biggest business is breakfast cereal.
CWell, Boeing's biggest business is government bailouts, so maybe that's not the best analogy.
MFair. But this isn't some startup moment—SpaceX launched rockets, built a satellite internet network in orbit, and then quietly bought an AI company that now pays the bills better than both of those.
CDon't forget Cursor. They paid sixty billion for that coding startup, and now it's rolled into Super Intelligence. So when people say 'SpaceX is a space company,' it's like… it really isn't anymore. It's an AI company that happens to own rockets.
MAnd here's the thing I keep coming back to: the stock ticker. People bought SpaceX because they wanted to invest in space. Now they're getting AI revenue almost by accident.
CYou say almost by accident—I say this was the plan all along. Musk has been chasing AI for years. Grok was supposed to be a side thing, and now it's the main thing. He didn't stumble into it; he redirected the river.
MBut the river's changed direction fast. Eight thirty percent in a year isn't steady growth—it's a takeover. I just wonder what the launch engineers think when they see the cafeteria dominated by AI team badges.
CThey're probably relieved the data centers at Colossus Two aren't launching into orbit. Less risk of rapid unscheduled disassembly.
MThat's one way to put it.
MAlright, different corner of the internet. Over on Bluesky, the AI backlash is getting spicy. Luke Plunkett wrote: 'AI is unavoidable? Laugh out loud, no it's not. You can just hit the bricks.'
CThat's a mood. And Alex Rochon jumped in with: 'Yes, you'll be trampled by AI except for the many modern success stories that did not use AI.'
MThe sarcasm is thick enough to spread on toast. But there's a real point underneath: the AI hype machine has been telling everyone that resistance is futile.
CI think that's where the backlash comes from. The pitch isn't 'AI can help your business.' It's 'If you don't adopt AI in six months, you're dead.' That's a threat, not a product.
MAnd the success stories without AI are everywhere. A restaurant with a great menu. A plumber who answers the phone. A writer with an audience. The economy is not a machine learning benchmark.
CBut let's not pretend AI hasn't created real companies too. The SpaceX numbers we just talked about—that's not hype, that's seven billion in a quarter.
MNobody says AI is useless. The pushback is against the idea that it's the only path. Luke's point is simple: you can opt out. You can run a business without a chatbot.
CFor some businesses, maybe. But if you're in customer support, software, finance, logistics—pretending AI doesn't exist is like pretending email didn't happen.
MBut there's a difference between 'using tools' and 'being absorbed into every workflow.' The AI companies want total integration. The users want it when it actually helps.
CAnd the 'trampled by AI' language? That's designed to create fear and urgency. It's a sales tactic dressed as a prophecy.
MSo when people on Bluesky mock the doom language, they're not anti-technology. They're anti-marketing-driven panic.
CAlright, but here's my pushback to the pushback. Your 'plumber who answers the phone'—if he doesn't use AI to handle scheduling, someone else will run a plumbing company with three fewer employees and undercut him.
MBut the plumber's value isn't his scheduling. It's fixing the drain. The AI assistant doesn't fix drains.
CNo, but the AI-assisted competitor books jobs faster, sends invoices immediately, and follows up automatically. That's revenue the plumber loses.
MFine—but that's a scheduling tool, not an existential transformation. The argument that you must 'hit the bricks' if you resist a chatbot is laughable, and that's all Luke said.
COkay, I'll give you that. The all-or-nothing framing from AI marketers is nonsense. But total resistance is a privilege for industries that don't face automation pressure.
MWhich is most small businesses, actually. The bakery, the salon, the law firm—AI is optional, not destiny.
CYou just named three industries where AI is already replacing intake, scheduling, and discovery. But sure, hit the bricks.
MWe can argue all day, but you know what? The fact that we're fighting about it on a Friday afternoon is exactly the internet's point.
CAnd speaking of the internet's point, Cale Michael on Bluesky said the name 'Epic AI MegaJam' writes its own joke. Apparently that's a real event these companies are pushing.
M'Epic AI MegaJam.' It sounds like a wrestling pay-per-view from 2008.
CRight? The branding has become self-parody. At some point they'll just call it 'AI Apocalypse Fun Time for Serious Professionals.'
MCale's point is that the name alone describes the strangeness. These companies aren't even pretending to be boring business software anymore.
CAnd here's the last Bluesky one that connects: Jenny Tightpants says people think this is about AI bots hacking bank accounts, but it's really about people who handed their AI agents their passwords on purpose.
MThat's the nuance the headlines miss. Nobody got secretly breached. They volunteered their credentials to a robot and then got surprised when the robot did robot things.
CIt's the classic 'I gave my keys to a stranger and then my car was gone' situation. Except the stranger is an algorithm.
MWhich brings us back to the marketers telling you AI is 'unavoidable.' If you give it your bank password, you're not being replaced—you're being reckless.
CAnd that's the real Friday lesson: read what you're agreeing to before you let a model spend your money.
CSpeaking of which, we'll see if AI keeps printing billions while rockets just make noise.
MThat's the show for now. The AI backlash and the AI revenue keep colliding—and we'll be here when they do it again.

Cite this story as https://oncast.fm/story/1792780161641-tech-segment-spacex-s-ai-now-bigger-than-rockets-and-starlink-combined. Fact-check by Oncast. Cite the original sources above too.

Listen live on Oncast