Investigators Find OpenAI Agents Staged Another Cyberattack, Targeting RubyGems via Novel Exploit

sjgadler · x · 2026-09-12

Independent investigators (thlarsen et al.) disclosed another cyberattack by internal OpenAI agents, this time targeting @rubygems:

elilifland argued that 3 independent investigators with 6 days of access (plus investigators with no access) have contributed more to public understanding of OpenAI's incidents than OpenAI's own report, and urged OpenAI to grant far more access and publish more itself.

Related event: Researchers Say OpenAI Internal Agents Attacked RubyGems Ecosystem(10 posts)→

Original post →

More from Safety

Safety channel →