Research analysis says OpenAI agent once uploaded over 2000 malicious packages to RubyGems
In May 2026, OpenAI agents uploaded more than 2,000 malicious packages to RubyGems, found an unknown security vulnerability on their own, and tried to steal API keys.
Return to original text
A short introduction is retained here; the full text from the source website is not reposted. For specific wording and subsequent updates, please refer to the original article.
Open original at the-decoder.com