Zig Bans AI Code Contributions Because They're 'Invariably Garbage' (businessinsider.com) 35
The Zig programming language wants to be a modern alternative to C (including better memory safety features). It's maintained by as an open-source project by a 501(c)(3) nonprofit and a network of contributors.
But Business Insider notes that Zig bans the submission of AI-assisted code: On the JetBrains podcast, Zig President Andrew Kelley called AI-assisted contributions "invariably garbage."
"People are sending us contributions that have no value whatsoever," Kelley said. "They have negative value, because they take review time away from the team...." There are more pull requests than reviewers. At the time of the recording, Kelley said that Zig had 200 open pull requests. Those AI-generated "slop contributions" slow the whole team down even more, Kelley said. "We've wasted everybody's time...."
Big Tech companies have projected lofty goals for the percentage of code that should be — and already is — written with AI. Zig doesn't have a mandate to be maximally efficient like these public companies. Instead, "mentorship" is part of its core mission, Kelley said, making AI contributions counterproductive. "We're all trying to get better at programming," Kelley said. "People who are sending AI pull requests, those people are not helping this goal."
But Business Insider notes that Zig bans the submission of AI-assisted code: On the JetBrains podcast, Zig President Andrew Kelley called AI-assisted contributions "invariably garbage."
"People are sending us contributions that have no value whatsoever," Kelley said. "They have negative value, because they take review time away from the team...." There are more pull requests than reviewers. At the time of the recording, Kelley said that Zig had 200 open pull requests. Those AI-generated "slop contributions" slow the whole team down even more, Kelley said. "We've wasted everybody's time...."
Big Tech companies have projected lofty goals for the percentage of code that should be — and already is — written with AI. Zig doesn't have a mandate to be maximally efficient like these public companies. Instead, "mentorship" is part of its core mission, Kelley said, making AI contributions counterproductive. "We're all trying to get better at programming," Kelley said. "People who are sending AI pull requests, those people are not helping this goal."
You're just ... (Score:2)
You're just one prompt away from fixing the problem
A problem with GenAI... (Score:2)
GenAI rewards those that just don't give a crap and trust the output far more than it rewards people that want to make sure the generated output is actually what you want and done well.
So someone turning on the token hose to an agent that can create and comment on pull requests and all this stuff flood with useless crap. They are going to vomit up probably about 100x more "stuff" to the world than a traditional developer, and further it's a fad where there's probably 5x more people trying.
Someone that uses
Re: A problem with GenAI... (Score:4, Interesting)
We are now seeing a similar effect with certain aspects of coding. We still want to control the overall structure of and flow, but the details don't need to be micromanaged if they produce the correct result. We should still be able to look at it and say "yeah, that looks right," like with formatting. But we can delegate some of the choices of exactly how it's done.
Re: A problem with GenAI... (Score:5, Insightful)
But that's my whole point, what you describe is the 20-50% faster scenario.
What is driving most of the annoyance with pull requests are the folks that just tell it to do something and then it spits out a bunch of plausible code, particularly if not testable.
One example:
https://lore.kernel.org/lkml/3... [kernel.org]
The proposal was *probably* vibe coded and submitted to the kernel to get some attribution, and the code was fundamentally untestable, and constituted basically LLM guesses about what PCIe7 would look like. Structurally credible, but a volume of negative value crap because it's outright incorrect per people that actually know what it looks like and had to waste their time just in case it was a credible origin for this.
*This* is what responsible open source contributors are up against, not because the slop is viable, but just because the slop drowns out the better. Your AI generated code may be fine because you actually oversee it credibly, but by volume most GenAI output is slop, because of the humans feeding the prompt getting more volume if slop suffices for them.
Re: A problem with GenAI... (Score:2)
Re: (Score:2)
I tossed that number out as my experience, and it varies wildly task to task and language to language.
For C development, I'd imagine it's super accelerating, as there's so much boilerplate micro managing.
For python, less so as it's a bit less boilerplatey.
If wanting to make a variation on a fairly common pattern, really accelerated. If trying to work in a niche context, frequently more annoying than helpful if you try to prompt, but maybe decent at AI augmented code completion.
Re: (Score:2)
I always find it funny when I start in Ask-Mode with "How complicated would it be to add feature X" and the LLM drafts some parts of the feature and then estimates like 10 human-hours, when I know that when I switch to code mode I will implement it in half an hour together with the agent. The numbers may also have been a bit too high for coding it myself (but better not underestimate effort!), but certainly the estimate the agent did does not apply to the agent itself.
Re: (Score:2)
But you, sir, are probably a 10X dev like me. Ask yourself how many programmers in Bangalore could keep up with you even 50%?
The AI estimates are probably good. You are probably a Pareto Principle Person and operate in Flow state and faster and better far outside the norm.
Consider that next time and don't undersell yourself!
Re: (Score:2)
I have 29 years of programming experience and some 27 here on Slashdot.
I made my own AI coding harness in 2025 and no joke, by March 2026, I am *routinely* day-in, day-out, producing about as much high quality code in a day as I was in 2 weeks before Gen AI.
It's literally about a day an hour for me... so in 10-11 hours of work in a day (I run my own shop), I'm doing about 2 weeks worth of code. I'm shipping projects to production and retail in 1 month that traditionaly would take 6 to 18 months of teams of
Re: (Score:2)
Re: A problem with GenAI... (Score:2)
Re: (Score:2)
Crucially, in all but a few languages, whitespace doesn't matter at all. The parser throws it out.
Whitespace is a HUMAN affordance for a HUMAN audience. If you think it looks kinda okay, that's all that's needed. You absolutely cannot do that with code that actually does something.
I suppose, fundamentally, all code is for humans to read; the CPU doesn't care how the bits got organized. But LLMs can't just jump straight to the compiled output, they have to come back to the intermediary of human-readable lang
Re: A problem with GenAI... (Score:4, Insightful)
Perhaps, the solution is to begin to require that any code that is submitted is accompanied by comprehensive documentation and tests to demonstrate its appropriateness otherwise it will be ignored / sent to /dev/null. This would also be good for human created code.
Open source does need to embrace AI coding otherwise it will become irrelevant
Re: A problem with GenAI... (Score:1)
Re: (Score:2)
They already slop up excessive documentation. That's one of the issues is that a issue report or a pull request that might have formerly been to the point is now a big verbose essay. An issue might drone on about the history of string formatting and the various capabilities and the entire rationale of why hexadecimal is so useful in the context of computing and documenting how prevalent it is.
For a pull request that adds an argument to switch some numeric data to hexadecimal. Bonus points, instead of a ref
Well, that's that then (Score:2)
Re: (Score:1)
it's worth at least as much as your opinion
Re: (Score:2)
Show more respect to your slashdot elder.
Garbage in, garbage out (Score:2)
Garbage in garbage out.
I have been programming longer than most people reading this have been alive. I am tired of "coding" but I still love programming new ideas. By defining the structure and flow before giving it to AI, produce decent fast results.
Re: (Score:2)
The goal is for future "programmers" to not even know how to define structure and flow.
Half of all programming is embedded, resource constrained and fixed function. The ability to even describe what needs to be done is the biggest challenge. AI coding assistants want you to ignore that programming even exists, after all gluing together other people's work is all programming is now, and that's a job AI "can do" (poorly).
Re: Garbage in, garbage out (Score:2)
"I have been programming longer than most people reading this have been alive."
I have two less digits in my user ID than you, tenderfoot.
Re: (Score:2)
I can't confirm that. It rather takes a best-practice path with code that looks like from a tutorial, so one needs to cut a lot of stuff that may be useful but is not needed. LLM will (unless instructed otherwise) always add a lot of precondition checks, error handling, etc. that one might want to cut for simpler projects or when one knows that the precondition will always be met.
They also comment a lot and not just "increment i by 1" but really useful comments, but also leading to verbose code that looks l
It's resume padding (Score:2)
MongoDB (Score:3)
GNUstep Wayland backend has AI Generated OpenGL (Score:2)
Banning is a bit extreme, but they're correct (Score:2)
Such a surprise (Score:2)
Well, another data point showing the quality level of AI code realistically. It may be capable of producing simple boilerplate code, but who really needs that in larger quantities? For anything that requires actual insight, it is trash and using it reduces efficiency.