SpaceX is making a bold move into artificial intelligence. Elon Musk’s rocket and satellite company announced a partnership with AI coding startup Cursor that includes an option to acquire the firm for $60 billion — a deal that signals SpaceX is no longer just a space company, but a serious AI contender gearing up for what could be the largest IPO in history.
The announcement came on Tuesday, with SpaceX declaring on X that it is working closely with Cursor to “create the world’s best coding and knowledge work AI.” The partnership merges Cursor’s software expertise with SpaceX’s powerful “Colossus” AI training supercomputer, with the goal of building what the company calls “the world’s most useful models.”
What the Cursor Deal Actually Looks Like
The deal’s structure is unconventional. SpaceX has secured the right to acquire Cursor later this year for $60 billion — or, if it decides not to buy, pay the startup $10 million for the work produced together. Tech and business observers have largely described it as a “try before you buy” arrangement, with a significant payoff for Cursor if the acquisition doesn’t go through.
Founded in 2022 and based in San Francisco, Cursor specializes in AI-powered coding tools, particularly for business applications. The company competes directly with Microsoft’s GitHub and faces growing pressure from OpenAI, which announced its coding tool Codex had reached four million weekly users, and Anthropic, whose Claude Code tool for developers has seen surging revenue.
The partnership gives Cursor access to SpaceX’s computing infrastructure — a lifeline the startup may urgently need. Right now, Cursor depends heavily on models from OpenAI and Anthropic to power its tools. Both of those companies are actively building their own competing coding products, which many observers say creates an existential platform risk for Cursor.
SpaceX and Cursor have also held discussions with AI company Mistral about a potential three-way partnership, according to reports.
Why the Tech World Is Buzzing
Reactions from investors, founders, and analysts have ranged from enthusiastic to skeptical.
Alex Finn, founder of AI startup Henry Intelligent Machines, called the deal a natural fit. He argued that xAI has lagged behind competitors on coding tools, while Cursor has a strong product but no model of its own. The partnership, in his view, addresses both weaknesses at once.
Tomasz Tunguz, general partner at Theory Ventures, framed it in terms of competitive layers. He said winning in agentic coding requires compute, models, and distribution — and that while xAI has enormous compute power through the Colossus data center in Memphis, it is losing popularity with developers. Cursor, he noted, has millions of users but is dangerously reliant on competitors for its model layer. “For $10 billion, SpaceX buys a call option on the distribution it couldn’t retain, and Cursor wins the independence it hasn’t yet secured,” he said.
Not everyone is convinced. Hadley Harris, cofounder of seed-stage VC firm Eniac Ventures, questioned the logic of the deal, saying that top developers he knows have already moved away from Cursor entirely.
Anand Kannappan, cofounder of PatronusAI and a former data scientist at Meta, offered a more nuanced take. He described the deal as a structured experiment — if training Cursor’s model on Colossus proves fruitful, SpaceX buys the company and owns the pipeline. If not, it pays $10 billion for the data and walks away. Either way, he argued, xAI comes out ahead.
SpaceX’s Bigger AI Ambitions
The Cursor deal fits into a much larger strategy. In February, SpaceX completed the acquisition of Musk’s xAI, a move Musk called “not just the next chapter, but the next book.” The merger was part of Musk’s stated plan to build solar-powered, satellite-based data centers to run future AI models in space.
Musk has argued that global electricity demand from AI cannot be satisfied with land-based infrastructure alone, and that moving computing into orbit — near vast, untapped solar energy — is the logical next step. SpaceX has already set the pace in the satellite market through Starlink, which operates the world’s largest satellite constellation.
However, SpaceX itself has noted in regulatory filings that unproven AI space data centers may not yet be commercially viable, a caution that has tempered some investor enthusiasm.
The IPO on the Horizon
Looming over all of this is SpaceX’s anticipated IPO. The company confidentially filed papers with U.S. regulators earlier this year, putting it on track for a potential listing by July. Media reports have pointed to a valuation of $75 billion or more for the offering itself, with a total market capitalization that could exceed $1.75 trillion — placing SpaceX among the ten largest companies in the world by market cap.
Analysts note that the IPO pitch is shifting. Investors are expected to evaluate SpaceX less on rocket launches and more on its “compute economics” — how efficiently it can convert AI infrastructure spending into profits. That’s a harder story to tell right now: capital expenditures more than doubled last year, surpassing revenue, and Starlink remains the primary cash engine funding SpaceX’s AI ambitions.
The pressure is real. If AI revenue doesn’t scale before costs escalate, SpaceX’s financial profile could start to look more like a cash-strapped startup than a trillion-dollar empire. For now, the Cursor deal is being watched closely as an early indicator of whether Musk’s sweeping AI bet will pay off — or prove to be one of the most expensive experiments in tech history.
