# Interesting LLM policy for the Zig language

**URL:** <https://discourse.haskell.org/t/interesting-llm-policy-for-the-zig-language/14007>\
**Category:** Links\
**Created:** [April 30, 2026, 1:04pm UTC](https://discourse.haskell.org/t/interesting-llm-policy-for-the-zig-language/14007 "2026-04-30T13:04:04Z")\
**Posts on this page:** 3\
**Page:** 1

<div class="post-metadata">

**Author:** ![LaurentRDC](https://sea2.discourse-cdn.com/flex002/user_avatar/discourse.haskell.org/laurentrdc/32/4950_2.png) [@LaurentRDC](https://discourse.haskell.org/u/LaurentRDC)\
**Post date:** [April 30, 2026, 1:04pm UTC](https://discourse.haskell.org/t/interesting-llm-policy-for-the-zig-language/14007/1 "2026-04-30T13:04:04Z")

</div>

For those who don’t read Hacker News 10x a day, here’s a discussion of the Zig project’s view of LLM contributions:

> **[The Zig project's rationale for their firm anti-AI contribution policy](https://simonwillison.net/2026/Apr/30/zig-anti-ai/)**
>
> Zig has one of the most stringent anti-LLM policies of any major open source project: No LLMs for issues. No LLMs for pull requests. No LLMs for comments on the …

The key quote (emphasis mine):

> Zig values contributors over their contributions. Each contributor represents an investment by the Zig core team - the primary goal of reviewing and accepting PRs **isn’t to land new code, it’s to help grow new contributors who can become trusted and prolific over time**.
> 
> LLM assistance breaks that completely. It doesn’t matter if the LLM helps you submit a _perfect_ PR to Zig - the time the Zig team spends reviewing your work does nothing to help them add new, confident, trustworthy contributors to their overall project.

I like the idea of focusing on creating new contributors, but I’d like to offer another perspective.

In my experience maintaining [`beam`](https://github.com/haskell-beam/beam/), there’s been an influx of outside contributions. Some of these contributions have involved LLM assistance.  
`beam` is famously opaque to outside contributors (and even to me sometimes!), but I think that LLMs have given the confidence to potential contributors to take the step and help. Overall, the contributions I review are definitely high-quality from the start, so the social contract hasn’t been broken – no one has “shoveled” code into existence here.

All this to say: I like Zig’s notion of nurturing contributors, and I think some Haskell projects can _still_ benefit from LLM usage in the community due to the intrinsic complexity of some of the codebases.

---

<div class="post-metadata">

**Author:** ![jaror](https://sea2.discourse-cdn.com/flex002/user_avatar/discourse.haskell.org/jaror/32/3271_2.png) [@jaror](https://discourse.haskell.org/u/jaror)\
**Post date:** [April 30, 2026, 1:36pm UTC](https://discourse.haskell.org/t/interesting-llm-policy-for-the-zig-language/14007/2 "2026-04-30T13:36:07Z")

</div>

Even if there is some benefit in using LLMs, does it outweigh the costs? GHC’s GitLab crashed a few days ago due to the sheer volume of bots. I urge all users of LLMs to consider their negative externalities.

---

<div class="post-metadata">

**Author:** ![LaurentRDC](https://sea2.discourse-cdn.com/flex002/user_avatar/discourse.haskell.org/laurentrdc/32/4950_2.png) [@LaurentRDC](https://discourse.haskell.org/u/LaurentRDC)\
**Post date:** [April 30, 2026, 2:09pm UTC](https://discourse.haskell.org/t/interesting-llm-policy-for-the-zig-language/14007/3 "2026-04-30T14:09:16Z")

</div>

That’s a fair point. Even huge platforms like GitHub are struggling under the load