SpaceX Buys AI Coding Startup Cursor in $60 Billion Deal

On June 16, 2026, four days after SpaceX rang the opening bell on the largest IPO in stock market history, the company confirmed it was buying Anysphere, the startup behind the AI coding tool Cursor, for $60 billion in an all-stock transaction. Reported alone, that's a big number attached to a hot AI company. Looked at closely, it's a story about an option Elon Musk locked in months before anyone was watching, a four-year-old startup that grew faster than any software company on record, and a newly public company using its stock as currency four days after its shares started trading.

What SpaceX Actually Agreed To

The mechanics of the deal are more specific than the headline number suggests. Anysphere's common and preferred shareholders will convert their stock into SpaceX Class A common shares, with the exact exchange ratio set by the volume-weighted average of SpaceX's closing price over the seven trading days before the deal closes. SpaceX expects to close the transaction sometime in the third quarter of 2026, pending regulatory approval, at which point Cursor becomes a wholly owned subsidiary rather than a partner company.

The filing also disclosed the financial guardrails built into the agreement: a $10 billion termination fee if SpaceX walks away under specific conditions, and a separate $4 billion fee that kicks in if antitrust regulators block the deal outright. Those numbers alone tell you how seriously both sides are treating regulatory risk on a transaction this size.

From an MIT Dorm Project to the Fastest-Growing Software Company on Record

Anysphere was founded in 2022 by four MIT classmates who built Cursor as an AI-native alternative to traditional code editors, embedding AI assistance directly into the workflow rather than bolting on a separate chatbot. The growth curve since then is genuinely unusual even by AI industry standards. Cursor entered 2025 with an annualized revenue run rate around $2.5 billion. Its Series D round in November 2025, a $2.3 billion raise, valued the company at $29.3 billion. By February 2026, annualized revenue had already climbed past $2 billion in a narrower measure cited by Forbes, and by early June, ahead of the SpaceX deal, that figure had more than doubled again to roughly $4 billion, with about $2.6 billion of it coming from enterprise accounts rather than individual developers. Multiple outlets have described Cursor's climb as the fastest any business software company has grown to this scale.

At $60 billion, SpaceX is paying roughly 15 times Cursor's annualized revenue, one of the largest multiples ever paid for a software acquisition, and a clear signal that whoever ended up owning Cursor was going to pay a steep premium for it.

The Option Elon Musk Quietly Locked In Back in April

The June announcement wasn't the first move. On April 21, 2026, SpaceX had already secured an option agreement with Anysphere that gave it two paths forward: acquire the company outright for $60 billion later in the year, or walk away from full ownership and instead pay roughly $10 billion for a deeper collaboration and services arrangement. That structure meant SpaceX had effectively locked in Cursor as a strategic asset months before the public deal was announced, while leaving itself room to decide how deep to go once its own IPO gave it the stock to pay with.

At the time that option was struck, Cursor was in the middle of raising a separate $2 billion private round that would have valued the company above $50 billion, with Andreessen Horowitz, Nvidia, and Thrive Capital reportedly lined up to participate. SpaceX's eventual $60 billion offer came in above that private-market number, which likely made the decision easier for Anysphere's board.

Why SpaceX Wanted a Coding Startup, Not the Other Way Around

The strategic logic traces back to February 2026, when SpaceX completed its merger with Elon Musk's AI venture xAI in a deal that valued the combined entity at $1.25 trillion. That merger folded xAI's Grok chatbot, its Colossus supercomputer cluster in Memphis, and its broader AI research operation into SpaceX, turning what had been a rocket and satellite company into a three-legged business spanning space, connectivity, and AI infrastructure.

The problem was that xAI's AI coding capabilities lagged well behind Anthropic, OpenAI, and Microsoft's GitHub Copilot, despite Grok's broader consumer traction. SpaceX and Cursor had reportedly spent months building a joint AI coding model trained on Colossus infrastructure ahead of the announcement, intended to ship inside both the existing Cursor product and a new one called Grok Build. Buying Cursor outright, rather than just partnering with it, gives SpaceX two distribution channels for the same underlying technology: developers already inside the Cursor ecosystem, and the broader Grok user base xAI is trying to grow.

The Bidders Who Walked Away First

Cursor wasn't short on suitors before SpaceX closed the deal. According to CNBC, Microsoft had examined acquiring Anysphere at an earlier stage but ultimately chose not to submit a formal bid, a notable decision given that Microsoft already owns GitHub Copilot and had an obvious strategic reason to want Cursor off the market. Cursor's leadership had also rebuffed two separate acquisition approaches from OpenAI, prioritizing independence over a deal with one of the two companies whose frontier models power much of Cursor's own product.

That history matters context-wise. It suggests Anysphere's leadership wasn't shopping the company aggressively. SpaceX's option agreement, backed by the promise of a public listing and fresh stock to pay with, appears to have been the offer serious enough, and large enough, to change that calculus.

Four Days After the Biggest IPO in History

SpaceX's own path to the negotiating table is worth understanding on its own terms. The company priced its IPO at $135 a share on June 11, 2026, raising approximately $75 billion, the largest initial public offering ever recorded, and began trading on Nasdaq the next day at a valuation near $1.77 trillion. Shares opened around $161, a roughly 19% premium to the IPO price, and briefly pushed as high as $225.64 intraday on June 16, the same day the Cursor deal was announced, before pulling back in the sessions that followed.

Elon Musk told investors ahead of the listing that SpaceX had been cash-flow positive since around 2015 and that the IPO was meant to fund "a significant growth phase," citing plans for satellite expansion and space-based AI data centers. Buying a $60 billion company just four days after going public is about as clear a signal as a newly listed company can send about how it intends to use that capital, and it wasn't limited to SpaceX's traditional rocket and Starlink businesses.

What the Price Tag Says About AI Coding Tools' Real Economics

Stepping back from the deal mechanics, the $60 billion price is itself a data point about where the AI coding tools market actually stands. Cursor's revenue growth, from roughly $2.5 billion annualized at the start of 2025 to around $4 billion by mid-2026, happened alongside real strain: multiple outlets reported that Cursor had struggled with compute scale and growth constraints even as demand outpaced its ability to serve it efficiently. That's the tension sitting underneath a lot of fast-growing AI application companies right now: revenue is real and growing, but the compute costs behind serving that revenue grow just as fast, if not faster, and companies without their own infrastructure eventually need a partner who has it. SpaceX, sitting on Colossus and a freshly funded balance sheet, was exactly that kind of partner.

The Data Question Nobody's Fully Settled

One thread worth treating carefully rather than dismissing outright: investor Jason Calcanis has publicly alleged that Anthropic used large volumes of Cursor's user traffic to help improve its own coding models, a claim reported by NextBigFuture and attributed directly to Calcanis rather than confirmed independently. It's an allegation, not an established fact, but it points to a real structural question the SpaceX deal raises going forward: Cursor's own filings noted that its developer usage data could help improve xAI's models, including Grok. For a tool used daily by a large base of professional developers, what happens to that usage data, and who gets to train on it, is a legitimate governance question that outlasts the ownership change itself.

What This Actually Means If You Build on Cursor

For the developers and companies who've built their daily workflow around Cursor, the practical questions are less about the price tag and more about what changes underneath the product they already use. A tool that positioned itself as an independent, model-agnostic coding assistant, capable of running on top of models from Anthropic, OpenAI, or others, is now owned outright by a company that has its own frontier model to promote in Grok and Grok Build. Whether Cursor keeps genuine model choice for its users or gradually steers usage toward xAI's own models is the kind of decision that won't show up in a press release, but will show up in the product over the next year.

The deal still has to clear regulatory review before it closes in the third quarter, and a transaction this large, folding a dominant AI coding tool into a company that already owns a frontier AI lab, is a plausible candidate for antitrust scrutiny given how concentrated the AI coding market has already become between Microsoft, Anthropic, OpenAI, and now SpaceX. Until that review plays out, the honest read for anyone relying on Cursor professionally is to watch the product roadmap closely rather than assume nothing changes just because the code editor looks the same on launch day.