Hackaday

Open source maintainers are increasingly overwhelmed by AI-generated 'slop' pull requests submitted by autonomous agents. Jeff Geerling and Daniel Stenberg (curl) are among those raising alarms, with GitHub now offering options to disable PRs entirely or restrict them to invited collaborators only. While the root cause is human behavior—someone configured the agent—the flood of low-quality AI submissions is eroding the collaborative openness that made open source strong. Maintainers may be forced to close off their projects, sacrificing the serendipitous contributions from unknown developers that historically helped squash bugs.

What About The Droid Attack On The Repos?

<blockquote>
It’s now to the point that GitHub will let you turn off PRs completely, at which point you’ve given up a key piece of the ‘hub’s functionality.
</blockquote>
A common childhood illness of new technologies. Once again, the developer community is forced to find measures against their unreasonable or malicious use. But this time, the disease is more aggressive and dangerous, affecting the fundamental, vital principles of the open source software.
<blockquote>
It’s a problem for maintainers, to be sure, but it’s as much a human problem as an AI one. After all, someone set up that AI agent and pointed at your PRs.
</blockquote>
The logical key “impact” of the AI usage. But why not to setup the AI agent in the repo, wich will automatically review such PRs and drop/cancel AI slope ones automatically?

Vibe coding, using AI just for the sake of saying it, plastering AI everywhere, RAM/NVME/other parts prices skyrocketing…
It’s like we’re being punished AND we’re paying for it.
Maybe PR should come from “certified” devs that may or may not use AI? With a dev rating (like kudos or reputation).

Such a bad situation to be in. I’d lock PRs down and limit them to key people for sure. Or at least only allow signed PRs so slop authors were identified and blocked from creating issues and PRs altogether.

The fact that github had the audacity to give two of the worst possible options is just nuts. Couldn’t they give a better option? Here’s a recommendation for github, allow PRs only and only if the contributor is assigned to give that PR on an issue. Hey, it’s the simplest simple solution that even that hard on scale actually… or maybe it is I don’t know I don’t work for github.