If strike date comes and Cursor is in fact worth less than $60B... they can move to acquire it for that price. Or just let it "expire". And if it's worth more, they get a savage good deal. If the services were worth $8B anyway, it's hard to lose.
It seems less crazy to me through this lens. A straight acquisition, today, at $60B would in fact be crazy.
And their price is so high because it's markup on API rates. API rates, even without markup, are just insanely irresponsible for anyone to be spending on full-time daily usage.
They are catching up fast!
https://www.businessinsider.com/chamath-palihapitiya-ai-cost...
https://x.com/chamath/status/2029634071966666964
I suspect that as the value a company provides is more than its code, then increasing code churn does not lead to an equivalent increase in revenue. Even for a tech company, a business' concept, connections, knowledge, assets, non-coding staff, etc.. are a significant value and increasing code doesn't increase the throughput of that value. For non-tech companies code is the grease in the gears, not the gears themselves.
You also neglect that products like Cursor run on two margins, their own plus the API provider’s. That’s always going to come at a premium
The rates without a margin of profit (or with a negative one) are not real.
If you think this of users who use cursor then I don’t think you’ve used cursor much at all.
Only the foundation model companies offer cheap/subsidized compute.
If you're an app layer company, you're offering a 10x worse deal to your customers.
Foundation model companies are willing to lose money to win loyalty. Remains to be seen if it'll work.
So no it’s not an oxymoron.
I’ve also seen it mentioned a lot of people having 2, 3 or even more subscriptions, which I’m pretty sure that can easily go South when it comes to costs.
But, again, and the most important point, I’ve never seen a detailed post on what people spend on this AI thing on a monthly basis (let’s say).
Cursor makes a ton of money because the product is great. It's easily the most sophisticated harness out there, and it isn't an IDE anymore. It's an agent dashboard since version 3.
Suffice it to say it's not all idiot money being thrown at them by users.
I do have Copilot in VSCode and Cursor.
I thought both should be equal in solving problems - turns out Cursor with the same model selected somehow was able to solve tasks that Copilot would get stuck or run in loops.
They have some tricks on managing file access that others don’t.
If you had a really good (big) local model, maybe it's an option, but on the more common smaller (<32b) models, it will have similar problems in looping, losing context, etc. in my experience.
It's a nice TUI, but the ecosystem is what makes it good.
true, but its not worth $60 billion fucking quid.
the whole thing is driven by irrational stock market investers who NEED ai to be the thing that saves the world.
they're betting everything on it.
At some point it can be valued as a high growth business, the code that backs it is almost irrelevant if the business is strong.
This. They are after the harness engineering experience of the Cursor people, I'd assume the they want to absorb all that into Grok's offerings.
The value and the room for innovation on the harness side seems to be underestimated.
Oddly the harness also affects model training, since even GLM/Z.ai for example train (I suspect) their model on the actual Claude Code harness. So the choises made by harness engineers affects the model. For Kimi/Moonshot and OpenAI the company makes their own harness. Alibaba uses Gemini.
Very interesting dynamics.
Earning millions that way is much more complicated.
It is surprisingly easy to do it once someone else has done the work. Increasingly that's the nature of AI-based software engineering: point it at an existing tool and ask it to carefully duplicate features until it has parity. As you pointed out, frontier LLM companies happen to be well positioned to sell the resulting products.
But if someone frames it "engineering talent that knows how to make LLMs even better at software development than competition" it might.
I see with my own work it works so it is not like Devin that was basically a scam that was valued at 10 billion.
In this kind of context yeah feels like it is quite possible to be worth 60 billion.
AI coding is much more than just the model - all the tools that human use in IDE are also useful for AI. Claude Code on the other hand just works with grep.
I thought it was a Windows thing. My Windows work computer is so heavily managed and monitored I assumed that was why Copilot stops being able to get terminal output or find the file I'm looking at. It's the same problem in IntelliJ and VSCode, with different models trying to find things in different ways.
Now that I think of it though, I've only used Copilot at work. At home I use Debian but I've never tried using Copilot. Claude, OpenCode, Gemini, and IntelliJ's AI Chat pointed at local Ollama models never have issues finding files or reading files and terminal output.
IDEs are made for just a human to interact with code. I think the paradigm of forcing these tools that weren’t built for this to do this, is us trying to fit a square peg in a round hole.
Call me old, but don’t put ai in my ide. My ide was made for a human, not an ai. For the established players for sure it makes sense since they already have space on our machines. But for the new ones imo terminal, or dedicated llm interfaces are where it’s at.
If I’m writing code sure suggest the next line. If the machine is writing code, let it, and just supervise properly. and have the proper interface that allows the strength of each
Also shameless plug, I wrote something about this very thing. https://khalah.medium.com/getting-ai-to-work-right-27b750dba...
The way I work now in the Codex desktop app is that I spin up 3-5 conversations which work in their dedicated git worktree.
So while the agent works and runs the test suite I can come back to other conversations to address blockers or do verification.
Important is that I can see which conversation has an update and getting desktop notifications.
Maybe I could set this up with tabs in the Terminal, but it does not sound like the best UX.
Claude is still gold standard if you're not in an IDE though.
Cursor both lets you highlight specific lines multiple times per chat and is much quicker at finding stuff.
I wouldn't think so. At work I have both cursor and claude code and while I use both, cursor is by far the most pleasant to use. If I had to give one up, I'd let claude go.
you've formed an opinion on the value of the company without knowing how many users it has? Kind of proves my point, no?Reminds me of the famous dropbox post: https://news.ycombinator.com/item?id=9224 - I don't even know if dropbox still exists in 2026 but i'm still happily using rsync and mailing things around because dropbox has just absolutely never worked reliably for me, unlike my 2007 gmail account.
Likewise, if it were up to me, instagram and any business whose business model revolves around ads would be banned (because ads would be banned because advertisement is harmful in general).
Unsure how it would work in practice.
Massive understatement calling it "a not particularly good plugin". If it were that simple there wouldn't be a need to even do this.
yes. plus $2b ARR, 1m DAU
What they want is the massive user base, the data (Cursor has a lot of high quality coding data for training), the teams expertise in coding models and agents, and the Composer models
60 billion is a large number but these frontier labs are burning billions a month in compute alone, and SpaceX is IPOing soon so they'll have a lot of cash to spend
I'd say that a million users is pretty good, but 350.000 paying users isn't, if you're a $60B company. Someone else mentioned that Anysphere has a $1B ARR, but I seriously doubt that each user is forking over ~$3000 per year.
Why do you doubt $3k/yr? Corporate usage skews a lot higher, when it's evaluated against hiring, not as a nice to have addon.
If $10k/yr means you get work done with one less hire that's an easy decision.
You mean Kimi K2.5? They can get that for free.
Elon/xAI want Grok to become useful for coding. Cursor has enough data and expertise to create a useful coding model. They found a price and an arrangement that made sense for both parties.
SpaceX spending $1B a month on various AI services seems ~plausible
(EDIT - Or maybe it's an IP transfer, or maybe it's over a longer time horizon. Idk but SpaceX clearly expects value from 'our work together' even if they don't exercise.)
And on the AI development side they're the ones providing compute in the form of a "million H100 equivalent Colossus training supercomputer"... On top of the cash.
I didn't say it was Wise.
I said it seems within possibility for this, very particular, corporation.
But I agree that it's hard to articulate what Cursor services you could blow this much money on.
Maybe it is all just an option! Or maybe they get a bunch of IP either way?
The main frenzy with Cursor started when you could access Anthropic models practically for free.
Otherwise it is just VS Code.
This is a bit simplistic. It's the VS Code that everyone used before cc came to town. Real devs, on real projects. All that data they collected is worth a lot more than "just vscode". Their composer2 is better than kimi2.5 and it's just a finetune on that data.
xAI had a decent model in grok4 (it was even sota on a bunch of benchmarks for a few weeks), but they didn't have great coding models (code-fast was ok-ish but nothing to write home about, certainly nowhere near SotA). Now that they've been banned from using claude, they'll get their expertise + data to build a coding model on top of whatever grok5 will be + their cluster for compute.
It doesn't sound like a bad plan to me, financial shenanigans or not.
> If you enable “Privacy Mode” in Cursor’s settings: zero data retention will be enabled for our model providers. Cursor may store some code data to provide extra features. None of your code will ever be trained on by us or any third-party.
Note the "may store some code data" and "none of your code will ever be trained on". In general you never want to include actual customer code in training the data, because of leaks that you may not want. Say someone has a hash somewhere, and your model autocompletes that hash. Bad. But that's not to say you couldn't train a reward model on pairs of prompts + completions. You have "some code data" (which could be acceptance rate) and use that. You just need to store the acceptance rate. And later, when you train new models, you check against that reward model. Does my new model reply close enough to score higher? If so, you're going in the right direction.
> If you choose to turn off “Privacy Mode”: we may use and store codebase data, prompts, editor actions, code snippets, and other code data and actions to improve our AI features and train our models.
Self explainatory.
> Even if you use your API key, your requests will still go through our backend!
They are collecting data even if you BYOK.
> If you choose to index your codebase, Cursor will upload your codebase in small chunks to our server to compute embeddings, but all plaintext code for computing embeddings ceases to exist after the life of the request. The embeddings and metadata about your codebase (hashes, file names) may be stored in our database.
They don't store (nor need to store) plain text, but they may store embeddings and metadata. Again, you can use those to train other things, not necessarily models. You can use metadata to check if you're going in the right direction.
Cursor needs their own 1st party backend model.
Sounds like a match made in heaven.
OpenAI tried to acquire Windsurf last year for $3B and couldn't.
1) A gamble based on cursor's compute constraint 2) if 1) plays out, he can purchase cursor via shares of spaceX over valued shares, at a fixed price should the valuation increase.
Wild conjecture.
Trading billions worth of idle compute, in exchange for a high-strike call option on the #3 player in the most-promising-vertical for AI, plus (presmuably) some access to their data, starts to sound like not a bad trade. Especially if you're pre-committed to betting your entire rocket company on winning in AI, and you're currently in sixth or seventh place.
SpaceX has invested a small amount as a share of its value in XAI, and could survive the loss of its investment.
The cluster’s already paid for, so likely in the $2B range for operating cash needs. Not more than $5.
If I imagine bringing in Cursor’s team to build a frontier model, ideally combined with Grok, which has one of the few truly proprietary data feeds available to it, and with a much larger custom model Cursor can solidify a place, and I get to do a stock swap to buy it, this sounds like a bet worth making.
Upshot - I bet there’s an MS/oAI deal on IP on the back of this; meanwhile the cluster goes brrr.
not that it isn't wild regardless
I am starting to see some potential in moving back away from pure terminal, a mixed modality with AI. But it is not in the direction of IDE in any traditional sense.
I'd even go so far as to say that any competitors that are direct (windsurf, kiro, etc.) aren't even in the same universe. Cursor is just so much better, faster, has better features (plan and debug mode), and squeezes much better results/code out of the same models. They absolutely have some secrete sauce that the other options just don't have.
So for me it’s more of a data deal - Elon buying himself some insight into codebases and real dev usage patterns? Oh finally someone to use his dirty data centres
Cursor has 1B in enterprise revenue. It doesn't matter if people can clone their product, those deals don't move slowly
That' all well and good and they had astounding growth rates but doesn't mean much. And 1B in ARR is not _that_ much in comparison. Also, reportedly they spend all their revenue and they have no control over the spend-side. The models they use will very likely get much more expensive. All the foundation model companies have a competing product. Cursor has the first mover advantage, but that will only help then so much. There have been plenty companies who grew fast, had huge revenue, but failed in the end, because they never got profitable. That's also in the cards for Cursor, if they don't fundamentally change their business model
That said, people are increasingly migrating to CLI tools (Claude Code if you like the Claude models, Pi Agent if you want something that's highly customizable, Crush if you want something fun), or GUI tools that are less code-first (Codex GUI).
Totally not for everybody though. I can see why some people would hate it.
Add emotional hedges if needed but they are just emotional not financial.
Your argument is based on an assumption that cursor cannot lose value. Even if the market says it has.
No free lunch: an option is a bet for both sides. Zero sum.
This was a similar play for twitter by the same person
While an innovator at the time, today there are a lot of LLM coding solution, sold by model providers, model aggregators even open source ones , it’s not obvious what is being bought that isn’t a feature of vs code or one of the LLM agents ( as the dismissive saying goes )
Downside is capped (cost of services + deal structure), upside is asymmetric if Cursor outperforms.
That said, these deals always hinge on whether the “$8B in services” is real economic value or just internal accounting. If it’s the latter, the risk profile looks very different.
Not happening