Marco Ieni 🦀

All major open source projects are discussing or already discussed their AI policy: banning LLMs completely, accepting them, etc.
Is it too early to standardize these (similar to how we did for MIT, GPL, etc.) so that projects don't have to write their own AI policy from scratch?

Your Slack workspace needs a https://zizmor.sh emoji

I just realized we can't rewrite Rust in Rust since it's already written in Rust 😢

My new AI policy is working great 😂

I just received a PR named "main" from a bot user adding 1400 lines of codes with unchecked " I have read the AI policy".

Here's what the Rust Infrastructure team achieved in Q1 and what we want to do in Q2

https://blog.rust-lang.org/inside-rust/2026/04/14/infrastructure-team-q1-recap-and-q2-plan/

While preparing a talk I found this blog post from Google (Nov 25) sharing their experience in adopting Rust in the Android codebase.
Crazy numbers! 📉

"Instead of moving fast and then later fixing the mess, we can move faster *while* fixing things."

https://blog.google/security/rust-in-android-move-fast-fix-things/

Proof that I use the `—` symbol myself: I set up a Raycast snippet for it long ago.

Today I woke up with 9 PRs in release-plz opened by a first-time contributor. The PRs look like they were written by an agent with no human review or intervention.
I wrote an AI policy, hoping to reduce the spam: https://github.com/release-plz/.github/blob/main/AI_POLICY.md

⚠️ New VSCode default you might want to disable

Pi is the neovim of AI coding agents
https://youtu.be/fdbXNWkpPMY