Simon Willison · 2026-04-30 · notable
Simon Willison: The Zig Project's Rationale for Their Firm Anti-AI Contribution Policy
Simon Willison highlights Zig Software Foundation VP Loris Cro's defence of Zig's blanket ban on LLM-authored contributions: 'contributor poker' — review time spent on AI PRs doesn't grow the project's pool of confident, trustworthy human contributors.
Why Zig is one of four major projects (with NetBSD, GIMP, qemu) banning AI-generated patches outright.
Key specs
| Hn points | 516 |
|---|
What is it?
An April 30, 2026 post by Simon Willison summarising Loris Cro's argument for Zig's strict no-LLM contribution policy. Cro is VP of community at the Zig Software Foundation; the policy applies to the entire Zig codebase. Simon places it in context with The Consensus's recent survey of source-available projects, which counts only four big projects (Zig, NetBSD, GIMP, qemu) currently banning AI contributions outright.
How does it work?
Cro's framing is 'contributor poker.' If a maintainer spends review time on an LLM-authored PR, they're not building rapport with a human contributor who could grow into a long-term committer — and if the PR was mostly written by a model, the maintainer might as well have had their own LLM solve the problem and skipped the social transaction entirely. Zig's leadership also argued that GitHub's 'file an issue with Copilot' UX was making the policy harder to enforce, which is part of why the project moved its primary repo to Codeberg in late 2025.
Why does it matter?
The post lands as a concrete philosophical counterpoint to the AI-coding boom: Zig is rejecting the productivity story to protect a long-term trust and mentorship loop. With 500+ HN points and a thread by Simon — who's normally an LLM enthusiast — it's the clearest articulation yet of why some projects view AI contributions as a load-bearing cost on maintainer attention rather than free help.
Who is it for?
Open-source maintainers, AI policy watchers, Zig users