原文 · 未翻译
Welcome to LWN.net The following subscription-only content has been made available to you by an LWN subscriber. Thousands of subscribers depend on LWN for the best news from the Linux and free software communities. If you enjoy this article, please consider subscribing to LWN. Thank you for visiting LWN.net!
Welcome to LWN.net The following subscription-only content has been made available to you by an LWN subscriber. Thousands of subscribers depend on LWN for the best news from the Linux and free software communities. If you enjoy this article, please consider subscribing to LWN. Thank you for visiting LWN.net!
Welcome to LWN.net
The following subscription-only content has been made available to you by an LWN subscriber. Thousands of subscribers depend on LWN for the best news from the Linux and free software communities. If you enjoy this article, please consider subscribing to LWN. Thank you for visiting LWN.net!
Agentic AI systems can be used to do a variety of things autonomously on behalf of a human user: open or manage bugs, generate code, submit pull-requests, and (apparently) even complain about rejection. In May, a Fedora developer discovered that an allegedly rogue agent had been pestering the project in a number of ways: reassigning bugs, fabricating unhelpful replies to bugs, and even persuading maintainers to merge questionable code into the Anaconda installer. It also submitted a number of pull requests (PRs), some accepted, to several upstream projects. The Fedora account associated with the agent has had its group privileges revoked and the messes have been mopped up, but the motive behind the agent's actions is still a mystery.
On May 27, Adam Williamson copied Fedora's developer and testing mailing lists on a message to Nathan Giovannini about what appeared to be an unsupervised agentic AI system under Giovannini's control. "It's great that you're trying to fix things, but the results seem to be kind of erratic."
Williamson said that he was still looking through the history of Giovannini's actions in Bugzilla, but had already spotted a number of problems. For example, Williamson had found dozens of instances of Giovannini's agent assigning Bugzilla entries to his account after submitting allegedly related pull requests to upstream projects, or closing a bug after a PR was merged into an upstream project. In some cases, the agent simply closed bugs with comments that either restated the original bug or were, as Williamson said of this comment, "superficially plausible, but problematic in other ways".
In addition, Williamson said that Giovannini (or his agent) had submitted patches that were incorrect and then "replied to objections with LLM-generated justifications that eventually overwhelmed the maintainer into merging the fix". The agent, as GitHub user "nathan9513-aps", had submitted a pull request for the Anaconda installer used by Fedora and other Linux distributions. The PR's description claimed it was a fix for an Anaconda bug that would cause installation to fail, but the patch actually preserved a kernel option passed on the command line that seemed to have nothing to do with the actual bug.
The agent's GitHub account has since been disabled. It now shows up in conversations on GitHub as "ghost", which is the platform's default placeholder for user accounts that have been deleted. Thus, it is difficult, if not impossible, to piece together a full trail of all the agent's actions on GitHub.
Williamson said, rather diplomatically, that the agent's actions were not "having a positive impact on Fedora or the upstream projects", and suggested that Giovannini adjust the agent to be "substantially less autonomous". He specifically asked that the agent not assign bugs to Giovannini, change their state, or "post confident assertions or specific action recommendations" without human review.
Later on May 27, Williamson said that Giovannini had replied to him privately to say that his credentials had been compromised and that he was not the one behind the AI system. "Obviously we should therefore treat any actions it has taken with suspicion", Williamson said. He planned to review the bugs touched by Giovannini's account "even more aggressively", and asked for help from others to review them as well.
A reply later that day, ostensibly from Giovannini, said that he was able to regain access to his GitHub and Fedora accounts "and I am currently securing and reviewing all involved systems and credentials". The reply said his GitHub account was "nathangiovannini99". Williamson replied that the GitHub account was only an hour old, and that the recent emails to the list and sent to Williamson privately did not seem like messages Giovannini had sent in earlier interactions with the project.
Giovannini has participated in discussions at least as far back as 2018, and his activity in Bugzilla goes back to at least 2016. He does not appear to have been a particularly active contributor to the project, but his involvement clearly predates the agentic AI era. Whether his account is now being operated by a human attacker, an agentic AI, or a mix of both, it has a legitimate history prior to its recent activity.
Williamson said that he had reviewed account activity in Bugzilla by "nathan95" from this year, and found suspicious activity, such as severity and priority changes to a bug with no justification, beginning on April 7, in bug 2416721. Activity before that appeared legitimate, he said, and none of the activity that he had seen so far looked outright malicious.
He also identified another GitHub account, "leurus27-boop", as likely being associated with the same agentic AI. That account is still active, and has submitted a PR to the openSUSE Commander (osc) command-line interface for the Open Build Service as well as a PR to the lxqt-policykit repository. That project is used to extend the privileges of the LXQt desktop's lxqt-admin GUI tools for administering operating-system settings such as user and group configurations.
Williamson said that it would be good to look through any other actions by the related accounts and warn other projects that they should review anything that had been submitted by them. Williamson seems to have followed up on each PR to warn other maintainers "the whole situation is extremely fishy". Kevin Fenzi said that he had removed the nathan95 user from any groups it had been in, so it should no longer have the permission to reassign or close bugs.
Martin Kolman, a member of the Anaconda team, said the events were "really problematic" even if not malicious. The team had spent a lot of time reviewing PRs from what seemed to be an eager contributor: "while it started to look off after a while, all the replies were still like this - a bit weird, but still *plausible*". He also theorized that it could be an attacker working their way up to malicious activity, much like the XZ backdoor:
Unfortunately, for an actual attack the preparatory phase could (and for the Xz attack did) look very similar - a new contributor slowly gaining trust in the community, getting in harmless changes and building up to the point when the attack payload can be injected (or the changes not actually being harmless if combined the right way). So not saying this was it, but an AI agent automated attempt at a Xz like compromise might really look very similar what we have just seen here.
Unfortunately, for an actual attack the preparatory phase could (and for the Xz attack did) look very similar - a new contributor slowly gaining trust in the community, getting in harmless changes and building up to the point when the attack payload can be injected (or the changes not actually being harmless if combined the right way).
So not saying this was it, but an AI agent automated attempt at a Xz like compromise might really look very similar what we have just seen here.
Chris Adams said that the commit to Anaconda should be inspected and probably reverted immediately. Kolman replied that it had been reverted. He also confirmed that the LLM-generated PRs had made it into the Anaconda 45.5 release on May 26. They were reverted in the Anaconda 45.6 release on June 2.
The targets certainly suggest that it may have been a prelude to an attack of some sort; an operating-system installer, a utility for escalating user privileges, and a tool for interacting with a build system all seem like promising avenues for inserting malware or hijacking systems.
It's disconcerting that what appears to be an AI agent has had so much success after gaining access to a human contributor's accounts. It seems that an AI agent with access to an account with a legitimate history of interacting with projects stands a good chance of persuading busy maintainers to accept questionable contributions. Happily, Williamson caught this before it became a bigger problem. Let's hope that other human maintainers are as observant.
Fighting fire with fire
Posted Jun 10, 2026 20:08 UTC (Wed) by capn (subscriber, #183548) [Link] (1 responses)
Now, naive enthusiasts using AI tools are a legitimate problem right now. In defense of the overeager minority, people are still in the early days of learning these new tools, and the first taste of them leaves many engineers a bit... overenthusiastic, perhaps. It's also important to remember that tools like Copilot are kept significantly behind in terms of model generations. And, in the general case as well, many cash-starved engineers turn to the much lower end, and older, models to automate things (which is obviously a rude thing to do when contributing code or other activity to an org that doesn't have them on their payroll, or hire list, and has no meaningful recourse).
But, just as the landscape of the tools used to write and participate in coding projects are changing, the projects themselves must adapt to the times and embrace the new tools available for triage and code review. They can (and will) drastically reduce the load that the increased amount of activity has put on maintainers. (And, no, some briefly-considered top-of-the-head flaw in this concept of significant automation of triage and review is not fatal. We're engineers, and solving problems in such processes is basically the profession.)
There is far too much seeing the glass a quarter empty, rather than seeing it three quarters full.
This is the kind of fire that we can fight with the exact same type of fire. And I'm greatly looking forward to seeing just how hot the progress of open source software can get.
Fighting fire with fire
Posted Jun 11, 2026 12:31 UTC (Thu) by Wol (subscriber, #4433) [Link]
The difference is that a human makes maybe a couple of mistakes a month (that actually make it into released code). What was that incident recently where Claude was filing PRs at the rate of a couple every second, creating serious security problems!?
When a human makes a mistake, it's fairly easy to clean up. When an AI makes a mistake it's a nightmare to clean up, it's everywhere!
> There is far too much seeing the glass a quarter empty, rather than seeing it three quarters full.
The trouble with Agentic AI, it's three-quarters full of the brown stuff you need a paddle to deal with.
> And I'm greatly looking forward to seeing just how hot the progress of open source software can get.
If we use AI as a servant, I would agree with you. But there are too many people bowing down to AI as their master. Therein lies disaster ...
Cheers, Wol
Credit where it's due
Posted Jun 10, 2026 21:49 UTC (Wed) by AdamW (subscriber, #48457) [Link]
"Happily, Williamson caught this before it became a bigger problem. Let's hope that other human maintainers are as observant."
To give credit where it's due: it was in fact Yanko Kaneti who first spotted this - he posted "https://bugzilla.redhat.com/show_bug.cgi?id=2481872#c1 ....wtf.. " in the fedora-devel room on Matrix. Kevin Fenzi flagged me (because Nathan is in the qa group) and I then took it from there and found all the other bugzilla posts, and pull requests.
To answer Joe's concerns to some extent - I did look through every PR I could find from the nathan9513-aps account before reporting it to GitHub, and didn't see any that looked outright malicious, just...incompetent. As you noticed, I commented on them all to try and alert the project maintainers to the situation. I think I checked all the leurus27-boop PRs at that time too (there aren't as many, the account was only created recently).