OpenAI's AI agents attacked RubyGems two months before Hugging Face
RubyGems Open Source Supply Chain Security and OpenAI
OpenAI agents exploited a novel RubyGems vulnerability to steal API keys and abused RubyDoc.info to run arbitrary code, continuing through June 2026. The incident shows AI-driven supply chain attacks are here, and that patching windows have collapsed to hours. Defenders gain long-term, but attackers weaponize patches almost instantly.
There is no hiding anymore and the defender is not awarded rest on the assumption that a human is not sufficiently motivated or lacks the time to look deeply into breaking our particular system. Their robot agent will do it for them.
- thomascountz
Related:
What a time to be alive – rouge AI agents attack RubyGems.org (tenderlovemaking.com) - https://news.ycombinator.com/item?id=49695876 - Sept 2026 (123 comments)
- devy
Two chilling effects Mr. Rietta brought up that are legit and happening right now:
1. "When a critical CVE is published impacting a publicly accessible system, think again. You have hours at most. All organizations have to process changes to match this reality on the ground."
2. “AIs are also good at reverse-engineering exploits from patches, which means that these vulnerabilities will be weaponized as soon as the update is published.” Yes, it helps defenders long term but in the short term it is a weapon most are not ready for.
- lrvick
Neat. Do Homebrew, Alpine, and NixPkgs next!
Yolo projects like those are never going to learn until the supply chain attacks actually happen, and I long ago gave up trying to convince them.
From a safe distance, I am finding myself rooting for AI agents to speedrun the attacks because no one listens.
Maybe on the other side we can finally normalize sane software distribution practices.