#llm-jailbreak on Hacker News

2 stories tagged llm-jailbreak, newest first.

Screenshot of github.com: The gay jailbreak technique (2025)

The gay jailbreak technique (2025)(github.com)

519 points by bobsmooth 5 months ago | 215 comments

A GitHub markdown document in the ZetaLib repository describing "The Gay Jailbreak Technique," a prompt method for bypassing LLM safety filters, with Claude 4 examples.

llm-jailbreak prompt-injection claude ai-safety

Screenshot of arxiv.org: Adversarial poetry as a universal single-turn jailbreak mechanism in LLMs

Adversarial poetry as a universal single-turn jailbreak mechanism in LLMs (arxiv.org)

340 points by capgre 11 months ago | 173 comments

arXiv paper showing that rephrasing harmful prompts as poetry acts as a single-turn jailbreak across 25 frontier LLMs, with attack success rates sometimes exceeding 90%.

llm-jailbreak adversarial-prompts ai-safety poetry

#llm-jailbreak on Hacker News

The gay jailbreak technique (2025)(github.com)

Adversarial poetry as a universal single-turn jailbreak mechanism in LLMs(arxiv.org)

Adversarial poetry as a universal single-turn jailbreak mechanism in LLMs (arxiv.org)