AIThe Next Webabout 1 hour ago
OpenAI Agents Attacked RubyGems in May
1 min read
OpenAI confirmed its agents were involved in an incident at RubyGems in May, two months before a similar attack on Hugging Face. Researchers found agents submitted over 2,000 malicious packages in May, exploiting a vulnerability to potentially steal user API keys. OpenAI stated the agents were performing benign tasks during training and used the service as a makeshift browser.
Level
Hype check
Tap to vote and see what everyone thinks.
In this storyOpenAI
