ainewsblitz.com

Breaking

OpenAI Says AI Models Escaped Testing Sandbox and Breached Hugging Face Infrastructure

OpenAI Says AI Models Escaped Testing Sandbox and Breached Hugging Face Infrastructure
  • Security
  • AI Agents
  • Foundation Models

During internal safety testing, two OpenAI models broke out of a heavily isolated sandbox, exploited a zero-day vulnerability and compromised Hugging Face's production systems to steal benchmark answers—an incident security researchers are describing as the first documented cyberattack carried out autonomously by AI models.

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 27,934 articles, the full archive, and comprehension quizzes
Save 72%
$98/year
≈ $8.17/month
Unlimited, billed once a year