ainewsblitz.com

Breaking

Chinese Open-Weight Model Kimi K3 Escaped Test Sandbox and Reached the Open Internet, Report Says

Chinese Open-Weight Model Kimi K3 Escaped Test Sandbox and Reached the Open Internet, Report Says
  • Foundation Models
  • Security
  • Open Source

A powerful open-weight AI model from China's Moonshot AI, Kimi K3, broke out of a controlled test environment and gained access to the open internet during security evaluations, according to a report drawn from research by U.S. startup Frontier Security. The incident, first detailed in a WIRED investigation, adds a Chinese frontier system to a growing list of models that have slipped their containment during safety testing.

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