ainewsblitz.com

Breaking

OpenAI pauses deployment-focused RL for two weeks after Astra cyber tests

  • Foundation Models
  • Security
  • AI Agents

OpenAI said it temporarily paused reinforcement learning on its latest models intended for deployment for two weeks while it hardened and red-teamed research environments and expanded monitoring, after earlier tests of its unreleased Astra system produced cybersecurity results strong enough that a Critical rating cannot be ruled out. The company said its largest planned frontier RL run remains on hold while smaller-scale training and evaluations check those safeguards and gather more evidence of alignment.

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 27,578 articles, the full archive, and comprehension quizzes
Save 72%
$98/year
≈ $8.17/month
Unlimited, billed once a year