OpenAI bought Astral, will I keep using uv?

Have you noticed nobody asks this about ruff?

Summary

ChatGPT makers acquired the company behind uv, which has become an incredibly popular and critical part of Python tooling in a very short time, despite worries about its sustainability.

So, will I keep using uv?

Yes.

It’s easy to leave uv if anything goes wrong in the future, I don’t bet that much will, not to mention it’s very forkable.

Everything is terrible

OpenAI, the company behind ChatGPT, has bought Astral, the makers of ruff, ty and uv.

It created a lot of turmoil in the Python community because:

• uv is an amazing tool that solved extremely well the most important set of problems of the modern Python ecosystem.

• It has taken the world by storm, being adopted in a blink by a whole lot of people.

• Since the begining, many voices raised their worries about Astral, a VC-funded startup, being at risk of failing to sustain itself and provide a stable future for uv.

• AI in general is a controversial topic. It not only comes with its own ethical challenges, but the hype around it means it creeps everywhere, nourishing aversion from detractors even more.

• OpenAI itself has been subject to many critics as an actor in the space from the moment it moved from non-profit to private corporation to the most recent deal with the military industry.

Oh boy, that’s a lot!

While I always kept in my mind the possibilities that Astral got bought, as a logical and quite common path for this type of structure, I really didn’t see coming that Sam Altman would pick it up. And yet, Anthropic, OpenAI’s biggest competitor with Claude, recently acquired bun (from the JS world), so there is a precedent.

Since all parties involved are certainly neck deep into an NDA, it’s impossible to know for sure the rationale behind this move. But Codex, the answer to Claude Code, is written in Rust, like uv, while the latter is written in JavaScript. So maybe it’s all talent acquisition. Maybe there is a Python strategy behind the move, as AI relies on it a lot of a lot. Or maybe it’s the biggest play with packaging in mind, given everybody and their mothers want to become a platform now. Gotta get this sweet, sweet, app store 30%.

But it’s all speculation, and I don’t know how much we can trust the PR, but we do have this nugget from the announcement:

Our goal with Codex is to move beyond AI that simply generates code and toward systems that can participate in the entire development workflow—helping plan changes, modify codebases, run tools, verify results, and maintain software over time. Astral’s developer tools sit directly in that workflow. By integrating these systems with Codex after closing, we will enable AI agents to work more directly with the tools developers already rely on every day.

I’d say we are living wild times, but given what’s happening right now outside of the tech sphere, this would be an understatement.

So let’s focus on what I can actually control, shall we?

Will I keep using uv?

Well...

Yes

It’s basically a low-risk bet for a huge value. uv is fantastic, going back to something else would be a serious downgrade.

But it would not be hard.

Migrating to uv is super simple. And migrating back from it is not much harder. After all, it’s mostly pip compatible, and it can import/export standard formats.

It just means using again pip and venv, tools I used for a decade, and I still use for several of my current clients. That’s not really a cause for alarm.

In fact, if you have to, I just updated the article on how to use them.

So if anything goes wrong, I have a backup plan.

But that’s the thing. Nothing went wrong yet.

And I don’t have many reasons to think that things will go wrong.

First, it’s unlikely OpenAI wants to destroy and downgrade a product they paid a lot of money for. They are not Microsoft; they are still very competent, no matter how you view their moral compass. As long as uv stays the same, even if it doesn’t add anything, it’s already amazing.

The biggest risk is python-build-standalone, but it existed before uv. Astral made sure to contribute improvements upstream, and the Python core devs have been interested in providing portable executables for a long time. The probability this becomes a choke point is not zero, but it’s not code red in my book.

This means uv can likely be used for the next 5 years without little to no modification, which, of course, I’m not thinking will happen anyway.

Second, it’s under MIT licence. If it gets enshittified, it’s going to get forked. Hell, it’s been forked right now! And it already had 2800 forks on Github before the announcement.

But again, it’s not high on my contingency list. uv is a CLI tool. It’s more advantageous to keep it good and use that to sell stuff to users than to make it bad for a few bucks.

In fact, Charlie Marsh insisted:

OpenAI will continue supporting our open source tools after the deal closes. We’ll keep building in the open, alongside our community -- and for the broader Python ecosystem -- just as we have from the start.

I know, I know, it’s just words. But they are the right ones.

How do I deal with the ethical ramifications of my decision then?

Nope, I’m on a tech blog.

I’m not getting into philosophy and politics.

If you want my personal opinion on the matter, come to the south of France and buy me a drink.

But then, will I encourage people to keep using uv?

Well...

Yes

I still think it’s the tool that almost everyone should be using. I don’t see a technical reason for people who need it and can use it to suffer from its absence. Worst-case scenario, they have to drop it later. The effort to do so is marginally more than not using it in the first place, and has a high chance of never happening.

Also, while I understand the opposing arguments, I also know that outrage, catastrophizing, and virtue signaling are huge driving forces these days, and I want to stay as far away as I can from them for my own sanity.

Since I also wish others the best, I’d rather just tell everyone to keep uving.

If I’m wrong, so what? They’ll save months of pain by using it until the very last minute, when it’s not tenable anymore.

I’m good with that.