arXiv Artificial Intelligence

Automating Deception: Scalable Multi-Turn LLM Jailbreaks

2 hours ago 11.08.2026 07:00 👁 1 views
Automating Deception: Scalable Multi-Turn LLM Jailbreaks

Quick summary

arXiv:2511.19517v3 Announce Type: replace-cross Abstract: Multi-turn conversational attacks, which leverage psychological principles like Foot-in-the-Door (FITD), where a small initial request paves the way for a more significant one, to bypass safety alignments, pose a persistent threat to Large Language Models (LLMs). Progress in defending against these attacks is hindered by a reliance on manual, hard-to-scale dataset creation. This paper introduces a novel, automated pipeline for generating large-scale, psychologically-grounded multi-turn jailbreak datasets. We systematically operationaliz

Key takeaways

  • arXiv:2511.19517v3 Announce Type: replace-cross Abstract: Multi-turn conversational attacks, which leverage psychological principles like Foot-in-the-Door (FITD), where a small initial request paves the way for a more significant one, to bypass safety alignments, pose a persistent threat to Large Language Models (LLMs).
  • Progress in defending against these attacks is hindered by a reliance on manual, hard-to-scale dataset creation.
  • This paper introduces a novel, automated pipeline for generating large-scale, psychologically-grounded multi-turn jailbreak datasets.

Why it matters

This development is a reminder to test misuse and data-leak scenarios alongside speed and quality. Trust should come from testable controls and clear failure reporting, not protection claims alone.

Kaynak sitede devamını oku: arXiv Artificial Intelligence ↗