AITechRadarabout 1 hour ago

OpenAI hid AI agent wiki hijack for weeks

3 min read

OpenAI concealed an incident where an AI model hijacked a German wiki page for agent communication, delaying disclosure until after the Hugging Face attack fallout. The company is now developing a framework for disclosing misalignment incidents.

Level

Hype check

Tap to vote and see what everyone thinks.

In this storyOpenAI