Jailbreaking ChatGPT via Prompt Engineering: An Empirical Study

arXiv:2305.13860v2 Announce Type: replace-cross
Abstract: Large Language Models (LLMs), like ChatGPT, have demonstrated vast potential but also introduce challenges related to content constraints and potential misuse. Our study investigates three key research questions: (1) the number of different prompt types that can jailbreak LLMs, (2) the effectiveness of jailbreak prompts in circumventing LLM constraints, and (3) the resilience of ChatGPT against these jailbreak prompts. Initially, we develop a classification model to analyze the distribution of existing prompts, identifying ten distinct …

abstract arxiv challenges chatgpt constraints cs.ai cs.cl cs.se engineering jailbreak jailbreaking key language language models large language large language models llms misuse prompt prompts questions research study type types vast via