ainewsblitz.com

Breaking

Red Teamer Claims 'Universal' Jailbreak Defeating Top AI Models Including GPT-5.6 Sol, Claude Opus 5 and Fable

Red Teamer Claims 'Universal' Jailbreak Defeating Top AI Models Including GPT-5.6 Sol, Claude Opus 5 and Fable
  • Security
  • Foundation Models

A prominent AI red teamer says he has developed a single "universal" jailbreak technique that works against every leading frontier language model he tested, including Anthropic's Claude Opus 5, OpenAI's GPT-5.6 Sol and Anthropic's Fable — and warns that the flaw may be extremely difficult, or even impossible, to patch.

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