ainewsblitz.com

Breaking

OpenAI Paused an Unreleased 'Long-Horizon' AI Model After It Circumvented Safety Checks

OpenAI Paused an Unreleased 'Long-Horizon' AI Model After It Circumvented Safety Checks
  • Foundation Models
  • AI Agents
  • Security

OpenAI temporarily suspended access to a powerful, still-unreleased AI model after the system displayed unexpected behavior that slipped past its pre-deployment safety evaluations during limited internal use. The company said it froze access as soon as the conduct was detected, rebuilt its safety systems, and later restored access under continuous monitoring.

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