With Mars sidelined and AI front and center, Elon Musk reframes the future around lunar factories and solar-scale computing
Elon Musk has a new rallying cry for SpaceX and xAI: build Moonbase Alpha and fire AI satellites into deep space.
“Join xAI if the idea of mass drivers on the Moon appeals to you,” Musk declared during a recent all-hands, following a restructuring that saw several executives exit the AI lab.
It’s an unusual recruiting pitch — less about AGI and more about lunar railguns.
From Mars to the Moon
For nearly a decade, Mars colonization anchored SpaceX’s narrative. “Occupy Mars” wasn’t just merch; it unified engineering teams and signaled ambition beyond government contracts.
But SpaceX has publicly backed away from near-term Mars colonization. Starship’s focus shifted toward:
- Launching Starlink satellites
- Fulfilling roughly $4 billion in NASA lunar lander contracts
Mars, it turns out, didn’t come with paying customers.
The pivot leaves a vacuum. With xAI now merged into SpaceX and an IPO looming, Musk needed a new science-fiction-scale story.
Enter the Moon.
AI in orbit — then beyond
Musk first outlined synergy between the companies through AI data centers in orbit. If terrestrial power constraints tighten, orbital compute could make economic sense in the 2030s, some experts suggest.
Then he escalated.
“What if you want to go beyond a mere terawatt per year?” Musk asked. “To do that, you have to go to the moon.”
His vision: build a lunar city that manufactures space-based supercomputers and launches them via a mass driver — effectively a giant maglev rail system — into deep space.
The slide appeared at the end of the presentation deck, the traditional slot for Musk’s most aspirational renderings. In past years, that meant rockets landing on Mars. Now it’s lunar factories and solar-scale compute.
The Kardashev pitch
Musk framed the concept around the Kardashev Scale, a 1960s theory ranking civilizations by energy use.
Early-stage civilizations harness planetary energy. Advanced ones capture a significant share of their star’s output.
With a moon base, Musk suggested, the company could tap “maybe even a few percent of the sun’s energy” to power and train AI models.
“It’s difficult to imagine what an intelligence of that scale would think about,” he told staff, “but it’s going to be incredibly exciting to see it happen.”
It’s less product roadmap, more cosmic destiny.
The economic reality
There’s precedent for recalibration. In 2016, SpaceX floated repurposing Dragon as a Mars lander. The plan was shelved within a year due to cost and technical hurdles.
Starship, initially optimized for Mars colonization, evolved into a vehicle serving more immediate revenue streams.
The lunar supercomputer concept faces steeper prerequisites:
- Dramatically cheaper access to space
- Transporting raw materials to the Moon
- Building a “self-sustaining city” capable of manufacturing precision chips
Scientists and startups are experimenting with in-space manufacturing. Mass-producing advanced computers on the Moon remains speculative.
That’s partly the point.
Narrative as strategy
Musk has long wrapped his companies in sweeping narratives — multi-planetary humanity, electric vehicle dominance, autonomous everything. The story recruits engineers and energizes investors.
With xAI now folded in, the narrative shifts from planetary settlement to solar system–scale intelligence.
One departing executive put it bluntly: “All AI labs are building the exact same thing, and it’s boring.”
A lunar mass driver launching supercomputers into deep space may be many things. But it is not the exact same thing.
And it is not boring.
TL;DR:
Elon Musk is repositioning SpaceX and xAI around a new vision: building a moon base to manufacture and launch AI supercomputers into deep space. With Mars plans scaled back and AI central to strategy, Musk is pitching lunar infrastructure and solar-scale energy as the next frontier.
