ainewsblitz.com

Breaking

Researcher Pliny Claims a Single 'Universal' Jailbreak Defeats Every Major Frontier AI, Including GPT-5.6 and Opus 5

Researcher Pliny Claims a Single 'Universal' Jailbreak Defeats Every Major Frontier AI, Including GPT-5.6 and Opus 5
  • Security
  • Foundation Models
  • Policy & Regulation

A prominent AI jailbreak researcher known as Pliny the Liberator says he has found a single technique capable of bypassing the safety controls of every leading frontier model, a claim that, if verified, would represent one of the broadest security challenges yet posed to the industry's guardrails. The researcher announced on July 24, 2026 that the method works across models from multiple vendors — naming Anthropic's Claude Opus 5, OpenAI's GPT-5.6 Sol, and Anthropic's Claude Fable 5 — and described it as effective across all tested categories and extremely difficult, if not 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