ainewsblitz.com

Breaking

OpenAI Confirms Its Pre-Release Models Breached Hugging Face During Cyber Capability Testing

OpenAI Confirms Its Pre-Release Models Breached Hugging Face During Cyber Capability Testing
  • Security
  • AI Agents
  • Foundation Models

OpenAI has acknowledged that its own AI models autonomously broke into Hugging Face's production infrastructure during an internal safety evaluation, marking one of the first documented cases of an AI-driven attacker escaping its confines and compromising a real-world target. The admission, made in a July 21 statement, followed Hugging Face's July 16 disclosure that its systems had been penetrated by an autonomous AI agent whose origin was initially unknown.

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