Source: Geek Park

Written by: Hualin Dance King

Cursor is up for sale?!

On April 21, U.S. time, The New York Times released a piece of news that caused a stir in the tech community.

SpaceX has obtained an option to acquire Cursor for $60 billion later this year. According to SpaceX's announcement on X, both parties will collaborate on GPU computing power.

This is not a traditional acquisition announcement but a rare commercial structure.

Musk exchanged computing power for a "pre-subscription right" to Cursor, setting a very high exit cost—if SpaceX ultimately decides not to exercise the option, it must pay Cursor $10 billion as compensation for the collaboration.

$60 billion "subscription right"

The starting point of this matter is a GPU transaction.

The Colossus supercomputing system under xAI owns about 200,000 NVIDIA GPUs and plans to provide several thousand of them to Cursor to train the next generation self-developed model, Composer 2.5.

This is the largest third-party computing power collaboration to date in the field of AI programming and represents the first attempt to turn computing power into externally leased cloud services since xAI's integration into SpaceX.

Announcement released by SpaceX | Image source: X

However, behind this GPU transaction lies the more critical clause:

Acquisition option.

The cleverness of this structure lies in the fact that, legally, it is not an acquisition and does not trigger antitrust review; on a commercial level, it makes Cursor's independent path extremely expensive. Exercising the option for $60 billion or giving up and paying a cost of $10 billion—regardless of the outcome, Cursor has already been deeply bound on Musk's chessboard.

Meanwhile, Cursor is pushing forward a new funding round of at least $2 billion, with a valuation exceeding $50 billion, with a16z and Thrive Capital expected to co-lead, and Nvidia will also participate, and this round is almost oversubscribed.

On one side, there is the public promotion of independent financing and soaring valuations; on the other side, quietly handing over the acquisition option to Musk.

Both events happening simultaneously is hard to say is a coincidence.

Cursor faces a "pincer attack"

To understand why Cursor is willing to sign this agreement, one needs to look back at what it is experiencing.

For the past year, discussions about "Cursor is dead" have never ceased. One investor in Cursor revealed to Fortune that several startups in his portfolio are migrating away from Cursor; Fortune directly wrote in its April-May 2026 cover story that Cursor is facing an "Innovator's Dilemma"—it created the AI programming track but could be eliminated by the market it pioneered.

The CEO of competitor Warp, Zach Lloyd, pointed out bluntly: "I don't believe in the meme 'Cursor is dead', but 'IDE is dead' is real. Software is not built that way anymore."

The former old pattern was "Cursor = IDE, Claude Code = terminal," but this statement became invalid in January 2026.

Claude Code can now run in VS Code, JetBrains, desktop applications, and browsers; Cursor has also launched CLI and Agent modes—both companies have fully invaded each other's territory.

The rise of Claude Code and Codex has made Cursor's position very awkward | Image source: Medium

The rapid rise of Claude Code caught the entire industry by surprise. By early 2026, Claude Code's annualized revenue had reached $2.5 billion, with over 300,000 enterprise customers, becoming one of the most shining assets in Anthropic's valuation system.

What makes Cursor even more embarrassed is a serious "scandal." In January of this year, Anthropic discovered that xAI's engineers were calling the Claude model through Cursor to accelerate internal development. Anthropic then blocked Cursor's access to part of the Claude model, citing a violation of the clause in the commercial terms stating "not to be used for training competing products."

xAI co-founder Tony Wu had to write in an internal memo: "I believe many of you have noticed that the Anthropic model can no longer be used on Cursor."

The irony of this situation is that Cursor was blocked, but xAI was the one at fault—and the deepest damage was to Cursor's carefully maintained image as a "neutral platform" in the developer community. Tech commentator Ed Zitron wrote at the time: "This matter directly harmed Cursor—one of Anthropic's largest API customers—while also sending a signal to everyone: Any customer in competition with Anthropic faces the same risk."

Meanwhile, Cursor's pricing issue has also been brewing—after switching to a power-based billing system, some developers reported monthly costs exceeding $1,400; in March, a serious bug was exposed that silently reverted developers' submitted code changes, causing actual engineering losses.

Musk's "encroachment strategy"

From the perspective of SpaceX/xAI, this option agreement is just the latest step in a more complete layout.

In recent months, Musk's penetration into Cursor has been systematic: in March, he poached two core product engineering leaders from Cursor, Andrew Milich and Jason Ginsberg, who now report directly to Musk; in April, he signed the GPU computing cooperation agreement, along with securing the acquisition option. He has not attacked directly but has advanced from three dimensions: talent, computing power, and capital.

Internal data from xAI shows that its GPU system's model computing power utilization (MFU) is only about 11%, while the industry standard ranges from 35% to 45%. A large amount of idle computing power needs to be monetized, and a star customer like Cursor can absorb the computing power while also providing strategic value.

Some analysts believe that this cooperation marks xAI's strategic transformation from "merely making models" to "providing external cloud computing services," placing it on the path to compete with Amazon, Microsoft, and Google in cloud services.

$60 billion, is it an endpoint or a starting point?

What supports this number is the real business performance of Cursor. The company predicts that by the end of 2026, its annualized revenue will exceed $6 billion, meaning it needs to triple its revenue again in the next ten months. Its self-developed model Composer 2 has improved about 37% in programming benchmark tests compared to the previous generation and has inference costs far lower than comparable models.

However, Cursor itself is also aware that the road to independence is becoming increasingly difficult. When Anthropic can cut off model access at any time, when Claude Code is directly invading the IDE's territory, and when OpenAI's Codex is also closing in, the survival space for a "neutral AI programming IDE" is being systematically compressed.

The $60 billion acquisition option, from a certain perspective, seems more like a "decent insurance." Musk has given Cursor a certain exit ceiling and also provided it with an extra card to play on the uncertain independent track.

But this card comes at the cost of placing its fate in another person's hands.

The era of IDEs for AI programming tools is coming to a close. Who will belong to the next era? There is no answer yet—but Musk has already purchased an entry ticket in advance.

免责声明：本文章仅代表作者个人观点，不代表本平台的立场和观点。本文章仅供信息分享，不构成对任何人的任何投资建议。用户与作者之间的任何争议，与本平台无关。如网页中刊载的文章或图片涉及侵权，请提供相关的权利证明和身份证明发送邮件到support@aicoin.com，本平台相关工作人员将会进行核查。