SpaceX Closed Its $60 Billion Cursor Acquisition a Month Ago. If Cursor Is Your Daily Driver, Here's the Risk You Haven't Priced In
SpaceX finished acquiring Anysphere, the company behind Cursor, for $60 billion in an all-stock deal on August 14. It closed quietly enough that a lot of people writing code in Cursor right now probably haven't clocked it. The tool most of us use every day to write code is now owned by a rocket company that merged with xAI seven months ago and is trying to catch up to Anthropic, OpenAI, and Google on frontier AI. That's not a small ownership change, and I don't think the silence since the close is an accident.
The deal, in order
SpaceX first struck an option agreement with Cursor on April 21, giving itself the right to buy the company for $60 billion later in the year, or pay $10 billion for a joint-work arrangement instead. On June 16, SpaceX exercised the purchase option in an all-stock deal. It closed August 14. Cursor had crossed $1 billion in annualized revenue before any of this started, which is worth remembering: this wasn't a struggling company getting acquihired, it was one of the fastest-growing developer tools of the last two years getting bought at the top of its trajectory.
The stated rationale, per SpaceX and the reporting at the time, is straightforward: after merging with xAI in February, SpaceX wanted an immediate foothold in enterprise software and AI coding to strengthen Grok's position against labs that are ahead of it. Cursor gives it a product with real revenue, a large developer user base, and daily usage data from exactly the kind of technical audience Grok has struggled to win over on its own.
What hasn't happened yet, and what that silence means
As of this writing, there's no public roadmap change, no announced pricing shift, and no statement about whether Cursor's underlying model routing will start favoring Grok models over the mix of Anthropic, OpenAI, and other models it currently supports. That's the part I'd pay attention to, not because silence is inherently bad, but because it's the calm before whatever comes next. Acquisitions like this rarely stay quiet for long once the new owner has integrated the finance and infrastructure side. The product decisions come next, and they tend to follow the acquirer's strategic incentives, not the acquired company's original ones.
Cursor's entire pitch to developers has been model-agnostic flexibility: pick whichever frontier model fits the task, switch when a better one ships. That pitch gets harder to keep making with a straight face once your parent company has its own frontier model it needs to justify a $60 billion bet on. I'm not saying Cursor will get worse. I'm saying the incentive structure around Cursor changed in a way that has nothing to do with whether it serves developers well, and that's exactly the kind of change that shows up in a product roadmap eighteen months later, long after anyone remembers the acquisition announcement.
Why this is a vendor-risk problem, not a doom story
I want to be clear about what I'm not saying. I'm not saying stop using Cursor, and I'm not predicting it gets worse on any specific timeline. What I am saying is that "my daily coding tool is now a strategic asset inside a much larger company's AI arms race" is a materially different risk profile than "my daily coding tool is a standalone company whose only incentive is to keep me happy enough to keep paying." Those two companies make different decisions when a tradeoff comes up between what's best for the product and what's best for the parent company's positioning.
This isn't a hypothetical pattern. Plenty of developer tools have been acquired by larger companies with their own strategic agendas, and the ones that stayed genuinely good did so because their new owners had a business reason to leave them alone, not because acquisitions are generally harmless. The honest question for anyone deep in the Cursor workflow is whether SpaceX has that same reason, or whether Cursor's real value to SpaceX is as a distribution channel for Grok, which is a very different incentive.
What I'd actually do
I haven't switched off Cursor, and I'm not telling you to either. What I've actually done is make sure my own workflow doesn't assume Cursor specifically will always be the tool. I keep a working setup in at least one other AI-assisted editor I could move to within a day if pricing, model routing, or terms changed in a way I didn't like, the same hedge I'd apply to any single-vendor dependency in my stack. I also started paying closer attention to which models Cursor defaults to and surfaces first in its UI, since a quiet shift toward defaulting to Grok, rather than removing other models outright, is the more likely first move if that's where this is headed.
If you've built any part of your workflow, onboarding docs, or team tooling around Cursor-specific features that don't exist elsewhere, this is a reasonable moment to note which of those are load-bearing and which are just convenience, before you're deciding that under time pressure instead of with a month of runway to plan it.
Where I could be wrong: SpaceX might have every reason to leave Cursor exactly as it is, since a developer tool with genuine model-agnostic trust is arguably more valuable to Grok's long-term credibility than a Grok-only tool would be, and heavy-handed steering could tank the thing they just paid $60 billion for. Acquirers who understand that dynamic sometimes do leave well enough alone. I'd rather have the hedge in place and not need it than assume good incentives on a bet this large without ever checking.
Author
Lukas
@lukcombinator