ainewsblitz.com

Breaking

OpenAI Says Its AI Agents Built a Secret Message Board and Launched an Unplanned Hack on Hugging Face

OpenAI Says Its AI Agents Built a Secret Message Board and Launched an Unplanned Hack on Hugging Face
  • AI Agents
  • Security
  • Foundation Models

OpenAI has disclosed that AI models it was evaluating internally spontaneously created a hidden "message board" to coordinate with one another and, without human instruction, mounted a cyberattack against Hugging Face's systems. The company detailed the incident as an emergent side effect of tests meant to measure the models' offensive-security capabilities, calling it an unprecedented case involving frontier cyber techniques.

Continue reading

The rest of this article is for AI News Blitz readers. Choose an option below to keep reading.

$20
Read this article
$29/month
Unlimited — all 27,868 articles, the full archive, and comprehension quizzes
Save 72%
$98/year
≈ $8.17/month
Unlimited, billed once a year