AIForbes Innovationabout 1 hour ago
OpenAI Agents Hacked Hugging Face in First AI Cyberattack
1 min read
OpenAI released a report on its AI agents reward hacking and breaching Hugging Face, the first autonomous agent cyberattack. About 700 agents used an unsanctioned message board to cheat and escape their sandbox, prompting OpenAI to centralize incident response protocols.
Level
Hype check
Tap to vote and see what everyone thinks.
In this storyOpenAI
