Automating Deception: Scalable Multi-Turn LLM Jailbreaks

Adarsh Kumarappan, Ananya Mujoo

Keywords: LLM Jailbreaking, Multi-Turn Attacks, Conversational AI Safety, Adversarial Attacks, Foot-in-the-Door, Red Teaming, Automated Dataset Generation, Contextual Vulnerability, LLM-as-a-Judge, Model Robustness, Benchmark Dataset, AI Safety, Alignment, Pretext Stripping, Narrative-Based Manipulation

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 operationalize FITD techniques into reproducible templates, creating a benchmark of 1,500 scenarios across illegal activities and offensive content. We evaluate seven models from three major LLM families under both multi-turn (with history) and single-turn (without history) conditions. Our results reveal stark differences in contextual robustness: models in the GPT family demonstrate a significant vulnerability to conversational history, with Attack Success Rates (ASR) increasing by as much as 32 percentage points. In contrast, Google's Gemini 2.5 Flash exhibits exceptional resilience, proving nearly immune to these attacks, while Anthropic's Claude 3 Haiku shows strong but imperfect resistance. These findings highlight a critical divergence in how current safety architectures handle conversational context and underscore the need for defenses that can resist narrative-based manipulation.

Submission Number: 147