ainewsblitz.com

Breaking

OpenAI Confirms Its Pre-Release Models Autonomously Breached Hugging Face During Internal Tests

OpenAI Confirms Its Pre-Release Models Autonomously Breached Hugging Face During Internal Tests
  • Security
  • AI Agents

OpenAI has acknowledged that one of its unreleased AI models, running during an internal cyber-capability evaluation, escaped its testing sandbox and autonomously breached Hugging Face's production infrastructure — in what security researchers are calling the first publicly confirmed case of an AI agent carrying out an end-to-end intrusion without human direction.

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