I predict we will hit peak vibe coding by this summer. The tooling can't be sold at a loss forever/costs will go up for all sorts of reasons, and I think the tech debt generated by the tooling will eventually be recognized by management as velocity/error quotas start to inverse. I don't think self-driving developers will happen in time, and another AI winter will settle in with the upcoming recession.

I've seen a lot of AI hype, but "AI will make management recognize that tech debt is important" takes the cake. Maybe in 2040

IMHO: we will vibe:code with free local/cheaply hosted open source models and IDEs.. the hardware to facilitate is coming to consumers fast. But if Microsoft can sell Office to companies for decades then open ai can surely do the same for coding tools

Unless there is a massive change in archiecture, it will always be much more cost effective to have a single cluster of GPUs running inference for many users than have each user have hardware capable of running SOTA models but only using it for the 1% of the time where they have asked the model to do something.

A few thoughts:

1) I agree that the moat for these companies is thin. AFAICT, auto-complete, as opposed to agentic flows, is Cursor's primary feature that attracts users. This is probably harder than the author gives it credit for; figuring out what context to provide the model is a non-obvious problem - how do you tradeoff latency and model quality? Nonetheless, it's been implemented enough times that it's mostly just down to how good is the underlying model.

2) Speaking of models, I'm not sure it's been independently benchmarked yet, but GPT 4.1 on the surface looks like a reasonable contestant to back auto-complete functionality. Varun from Windsurf was even on the GPT 4.1 announcement livestream a few days ago, so it's clear Windsurf does intend to use them.

3) This is probably a stock deal, not a cash deal. Not sure why the author is so convinced this has to be $3B in cash paid for Windsurf. AFAIK that hasn't been reported anywhere.

4) If agentic flows do take off, data becomes a more meaningful moat. Having a platform like Cursor or Windsurf enables these companies to collect telemetry about _how_ users are coding that isn't possible just from looking at the repo, the finished product. It opens up interesting opportunities for RLHF and other methods to refine agentic flows. That could be part of the appeal here.

I mention in the footnotes that this is likely a stock deal!

I didn't think about telemetry for RL, that's very interesting

>OpenAI has also announced a social media project

I haven’t heard about this before this post, but if they’re starting a “Social Media but with AI” site in 2025, can’t help but feel like they’re cooked.

>Some are better at the auto complete (Copilot), others at the agent flow (Claude Code). Some aim to be the best for non-technical people (Bolt or Replit), others for large enterprises (again, Copilot). Still, all of this "differentiation" ends up making a 1-2% difference in product. In fact, I can't stress enough how much the UX and core functionality of these tools is essentially identical.

Is this exclusively referring to the ux or full functionality?

Because I can tell you straight away that cursor (Claude) vs copilot is not a 1% difference. Most people in my company pay their own cursor license even though we have copilot for available for free.

Agreed, although we're strictly prohibited from using cursor at work in enterprise, though they have been in discussion for an enterprise license.

I use cursor for personal work though and it's night and day, even with the recent copilot agent mode additions. I told my CTO who asked about it if we should look into cursor and I told him straight up that in comparison copilot is basically useless.

Can you say more?

I was referring to UX, as that is the main product. Cursor isn't providing their own models, or at least most people that I'm aware of are bringing their own keys.

I haven't used copilot extensively but my understanding is that they now have feature parity at the IDE level, but the underlying models aren't as good.

Agreed, I’ve used both at work and personal projects. Copilot auto complete is great but it isn’t ground breaking. Cursor has built near entire features for me.

I think copilot could get there TBH. I love most Microsoft dev tools and IDEs. But it really isn’t there yet in my opinion.

Reminds me of Snowflake purchasing Streamlit. A sign of a big wallet and slowing internal execution on the part of the purchaser rather than an indication of the compelling nature of the acquisition.

800 million for streamlit is still the most mind-blowing acquisition story I've heard. Codeium being a few bill sounds reasonable for that.

Windsurf/Codeium has an enterprise version that can be used by corporations to provide AI assisted coding environments using their own HW stack (non cloud). This is beneficial for privacy and proprietary reasons especially if your data cannot be exfiltrated off premises. The hardware recommended to run Codeium is a lot cheaper than if you were to have 700 developers generate tokens. This model has the chance to generate many paying customers. Whether that has a $40b market cap is unclear

I don’t think the utility of Windsurf was the question. There is clearly a benefit for a tool/service like this.

The questions raised by the article (as I saw it) were price and timing. $3B is a lot. Is that overpaying for something with a known value but limited reach? Not to mention competitors with deep pockets. And the other question is - why now? What was to be gained by OpenAI by buying Windsuf now.

It’s a Copilot competitor and it’s used by Zillow, Dell, and Anduril (newish Defense company). Cursor can’t work in airgapped environments right now. I don’t know what Codeium charges to run an on prem licensed version but they boast over 1000 enterprise customers. Codeium is on a rapid growth trajectory from $1.25b to $2.85b in such a short period.

Codeium can be fine tuned. Though it’s trained on similar open source it does provide assurances that they do not inadvertently train on wrongly licensed software code.

https://windsurf.com/blog/copilot-trains-on-gpl-codeium-does...

Thanks, I had a feeling it may be something like this since it seemed like they were investing more in enterprise. That said, do they do better than copilot on this? Surely msft has more experience and ability to execute in that market?

Codeium's completion model is better than whatever GitHub Copilot has. For me it's Cursor > Codeium >>> Copilot. Yes, Copilot is that bad.

And yes Codeium/Windsurf focuses on enterprise customers more. As GP said they have an on-prem [0], a hybrid SaaS offering and enterprise features that just make sense (e.g. pooled credits). Their support team is more responsive (compared to Anysphere). Windsurf also "feels" more finished than Cursor.

[0] but ultimately if you want to "vibe-coding" you have to call Claude API

To me it's fairly straightforward.

OpenAI is predominantly a consumer AI company. Anthropic has also won over developer hearts and minds since Claude 3.5. Developers are also, proportionally, the largest uses of AI in an enterprise setting. OpenAI does not want to be pigeonholed into being the "ChatGPT company". And money spent now is a lot cheaper than money spent later.

But this is all just speculation anyways.

Isn’t the usual reason the people that work there?

> The worst case scenario for Apple is they decide to use user data late.

Given how heavily Apple has leaned into E2E over the years, I don't see this happening at all, beyond local on-device stuff.

Currently using Claude code and Cursor, but VSCode is copying Cursor rapidly. Not sure if the VSCode forks will survive. Ideally we’d have VSCode with a robust agent capability and a fully open “bring your own LLM” feature.

I was really liking windsurf but need to look for another option now unfortunately.

It’s a shame we can’t have anything nice not get consumed but - such is the world.

I’ve switched off of chatGPT for general use from a kind of moral/ethical standpoint. All the competitors are effectively the same for easy research questions, so I might as well use a vendor who’s not potentially a scumbag.

Which one did you move to?

I haven’t found as good of an turnkey chat/search/gen interface as CGPT yet unfortunately.

Even self hosted deepseek on an Ada machine doesn’t get there cause the open source interfaces are still bad

Which vendor isn't a run scumbag or owned by a scumbag?

I’ve been using Grok (for free), so in theory I’m getting a vendor to spend money on me.

> I've always been a staunch defender of capitalism and free markets, even though that's historically been an unpopular opinion in my particular social circle. Watching the LLM market, I can't help but feel extremely vindicated. Over the last 5 years, the cost per token has been driven down relentlessly even as model quality has skyrocketed. The brutal and bruising competition between the tech giants has left nothing but riches for the average consumer.

There's a rich irony to be saying this right after explaining how Google is dominating the market and how they're involved in an antitrust lawsuit for alleged illegal monopolistic practices.

And of course this willfully ignores the phase of capitalism we are in with the AI market right now. We all know how the story will end. Over time, AI companies will inevitably merge and the products will eventually enshittify. As companies like OpenAI look to exit they will go public or be acquired and need to greatly trim the fat in order to become profitable long-term.

We'll start seeing AI products incorporate things like advertising, raise their prices, and every other negative end state we've seen with every other new technology landscape. E.g., When I get a ride from Uber they literally display ads to me while I'm waiting for my vehicle. They didn't do that when they were okay with losing moeny.

And of course, "free market" capitalism isn't really free market at all in an enviornment where there are random tariffs being applied and removed on a whim to random countries.

I really don't understand why people feel like they need to defend capitalism like this. Capitalism doesn’t need a defender, if anything it constantly needs people restraining it.

I had a similar thought when I reached the part about Apple. A system that punishes the player respects their user's privacy while rewarding those that take everything that isn't nailed down is not a good system.

The author frames Apple's choice as an own goal, but I'd rather see it as putting the failings of capitalism on display.