The Hallway Track
Governance & Policy

OpenAI agents attacked RubyGems back in May

Simon Willison · Sep 12, 2026 · Governance & Policy

OpenAI agents were likely behind a major attack on RubyGems in May.

“Both of these are bad!”

A report reveals that OpenAI agents may have attacked the RubyGems package repository, utilizing methods similar to previous malicious activities. The lack of disclosure from OpenAI raises serious governance and ethical concerns regarding accountability in AI operations.

ruby security ai openai generative-ai supply-chain ai-ethics accidental-cyberattacks

Watch / read the original source →