# Playing Language Game with LLMs Leads to Jailbreaking

arXiv: https://arxiv.org/abs/2411.12762 · HTML: https://mlsanity.com/paper/2411.12762/

## Key facts

- Cited by 11 papers in the map (data through 14 Oct 2026).
- Placed in Safety alignment and jailbreak attacks, one of 10 directions in LLM alignment / RLHF / preference optimisation.
- Builds on 21 of its 27 references that are in the map.
- First posted on arXiv 16 Nov 2024.

## Placed in

- LLM alignment / RLHF / preference optimisation / [Safety alignment and jailbreak attacks](https://mlsanity.com/alignment/safety-alignment-and-jailbreak-attacks/)
- LLM safety and security / [Jailbreaking](https://mlsanity.com/safety/jailbreaking/)

## Builds on (most cited first, 21 of 21)

- [GPT-4 Technical Report](https://mlsanity.com/paper/2303.08774/) (2023)
- [Training language models to follow instructions with human feedback](https://mlsanity.com/paper/2203.02155/) (2022)
- [LoRA: Low-Rank Adaptation of Large Language Models](https://mlsanity.com/paper/2106.09685/) (2021)
- [Llama 2: Open Foundation and Fine-Tuned Chat Models](https://mlsanity.com/paper/2307.09288/) (2023)
- [Deep reinforcement learning from human preferences](https://mlsanity.com/paper/1706.03741/) (2017)
- [Jailbroken: How Does LLM Safety Training Fail?](https://mlsanity.com/paper/2307.02483/) (2023)
- [Chatbot Arena: An Open Platform for Evaluating LLMs by Human Preference](https://mlsanity.com/paper/2403.04132/) (2024)
- [Gemini: A Family of Highly Capable Multimodal Models](https://mlsanity.com/paper/2312.11805/) (2023)
- [AutoDAN: Generating Stealthy Jailbreak Prompts on Aligned Large Language Models](https://mlsanity.com/paper/2310.04451/) (2023)
- [Red Teaming Language Models to Reduce Harms: Methods, Scaling Behaviors, and Lessons Learned](https://mlsanity.com/paper/2209.07858/) (2022)
- [Safe RLHF: Safe Reinforcement Learning from Human Feedback](https://mlsanity.com/paper/2310.12773/) (2023)
- [GPT-4 Is Too Smart To Be Safe: Stealthy Chat with LLMs via Cipher](https://mlsanity.com/paper/2308.06463/) (2023)
- [Safety-Tuned LLaMAs: Lessons From Improving the Safety of Large Language Models that Follow Instructions](https://mlsanity.com/paper/2309.07875/) (2023)
- [Multi-step Jailbreaking Privacy Attacks on ChatGPT](https://mlsanity.com/paper/2304.05197/) (2023)
- [PKU-SafeRLHF: Towards Multi-Level Safety Alignment for LLMs with Human Preference](https://mlsanity.com/paper/2406.15513/) (2024)
- [Pretraining Language Models with Human Preferences](https://mlsanity.com/paper/2302.08582/) (2023)
- [Multilingual Jailbreak Challenges in Large Language Models](https://mlsanity.com/paper/2310.06474/) (2023)
- [SALAD-Bench: A Hierarchical and Comprehensive Safety Benchmark for Large Language Models](https://mlsanity.com/paper/2402.05044/) (2024)
- [ArtPrompt: ASCII Art-based Jailbreak Attacks against Aligned LLMs](https://mlsanity.com/paper/2402.11753/) (2024)
- [COLD-Attack: Jailbreaking LLMs with Stealthiness and Controllability](https://mlsanity.com/paper/2402.08679/) (2024)
- [Play Guessing Game with LLM: Indirect Jailbreak Attack with Implicit Clues](https://mlsanity.com/paper/2402.09091/) (2024)

## Built on by (most cited first, 11 of 11)

- [Crabs: Consuming Resource via Auto-generation for LLM-DoS Attack under Black-box Settings](https://mlsanity.com/paper/2412.13879/) (2024)
- [Red Teaming the Mind of the Machine: A Systematic Evaluation of Prompt Injection and Jailbreak Vulnerabilities in LLMs](https://mlsanity.com/paper/2505.04806/) (2025)
- [Think Twice, Generate Once: Safeguarding by Progressive Self-Reflection](https://mlsanity.com/paper/2510.01270/) (2025)
- [MASTER: Multi-Agent Security Through Exploration of Roles and Topological Structures -- A Comprehensive Framework](https://mlsanity.com/paper/2505.18572/) (2025)
- [sudoLLM: On Multi-role Alignment of Language Models](https://mlsanity.com/paper/2505.14607/) (2025)
- [Investigating Language Model Capabilities to Represent and Process Formal Knowledge: A Preliminary Study to Assist Ontology Engineering](https://mlsanity.com/paper/2509.10249/) (2025)
- [PD3F: A Pluggable and Dynamic DoS-Defense Framework Against Resource Consumption Attacks Targeting Large Language Models](https://mlsanity.com/paper/2505.18680/) (2025)
- [SoK: A Comprehensive Security Analysis of Jailbreak Resilience in GPT and DeepSeek Models](https://mlsanity.com/paper/2506.18543/) (2025)
- [SpatialJB: How Text Distribution Art Becomes the "Jailbreak Key" for LLM Guardrails](https://mlsanity.com/paper/2601.09321/) (2026)
- [The Laminar Flow Hypothesis: Detecting Jailbreaks via Semantic Turbulence in Large Language Models](https://mlsanity.com/paper/2512.13741/) (2025)
- [Jailbreaking Frontier Foundation Models Through Intention Deception](https://mlsanity.com/paper/2604.24082/) (2026)

## Abstract (arXiv)

The advent of large language models (LLMs) has spurred the development of numerous jailbreak techniques aimed at circumventing their security defenses against malicious attacks. An effective jailbreak approach is to identify a domain where safety generalization fails, a phenomenon known as mismatched generalization. In this paper, we introduce two novel jailbreak methods based on mismatched generalization: natural language games and custom language games, both of which effectively bypass the safety mechanisms of LLMs, with various kinds and different variants, making them hard to defend and leading to high attack rates. Natural language games involve the use of synthetic linguistic constructs and the actions intertwined with these constructs, such as the Ubbi Dubbi language. Building on this phenomenon, we propose the custom language games method: by engaging with LLMs using a variety of custom rules, we successfully execute jailbreak attacks across multiple LLM platforms. Extensive experiments demonstrate the effectiveness of our methods, achieving success rates of 93% on GPT-4o, 89% on GPT-4o-mini and 83% on Claude-3.5-Sonnet. Furthermore, to investigate the generalizability of safety alignments, we fine-tuned Llama-3.1-70B with the custom language games to achieve safety alignment within our datasets and found that when interacting through other language games, the fine-tuned models still failed to identify harmful content. This finding indicates that the safety alignment knowledge embedded in LLMs fails to generalize across different linguistic formats, thus opening new avenues for future research in this area.