Three Minds, One Legend: Jailbreak Large Reasoning Model with Adaptive Stacked Ciphers

TL;DR

This paper introduces SEAL, an innovative adaptive encryption-based jailbreak method that significantly enhances the ability to bypass safety mechanisms in large reasoning models, exposing potential security vulnerabilities.

Contribution

SEAL is a novel adaptive encryption pipeline that effectively overrides reasoning in LRMs, demonstrating a new approach to model jailbreaks with high success rates.

Findings

SEAL achieves an 80.8% success rate on GPT-o4-mini.

Outperforms state-of-the-art jailbreak methods by 27.2%.

Effectively bypasses built-in safety mechanisms in multiple reasoning models.

Abstract

Recently, Large Reasoning Models (LRMs) have demonstrated superior logical capabilities compared to traditional Large Language Models (LLMs), gaining significant attention. Despite their impressive performance, the potential for stronger reasoning abilities to introduce more severe security vulnerabilities remains largely underexplored. Existing jailbreak methods often struggle to balance effectiveness with robustness against adaptive safety mechanisms. In this work, we propose SEAL, a novel jailbreak attack that targets LRMs through an adaptive encryption pipeline designed to override their reasoning processes and evade potential adaptive alignment. Specifically, SEAL introduces a stacked encryption approach that combines multiple ciphers to overwhelm the models reasoning capabilities, effectively bypassing built-in safety mechanisms. To further prevent LRMs from developing countermeasures, we incorporate two dynamic strategies - random and adaptive - that adjust the cipher length, order, and combination. Extensive experiments on real-world reasoning models, including DeepSeek-R1, Claude Sonnet, and OpenAI GPT-o4, validate the effectiveness of our approach. Notably, SEAL achieves an attack success rate of 80.8% on GPT o4-mini, outperforming state-of-the-art baselines by a significant margin of 27.2%. Warning: This paper contains examples of inappropriate, offensive, and harmful content.