Benchmarking and Defending Against Indirect Prompt Injection Attacks on Large Language Models

arXiv:2312.14197v2 Announce Type: replace
Abstract: The integration of large language models (LLMs) with external content has enabled more up-to-date and wide-ranging applications of LLMs, such as Microsoft Copilot. However, this integration has also exposed LLMs to the risk of indirect prompt injection attacks, where an attacker can embed malicious instructions within external content, compromising LLM output and causing responses to deviate from user expectations. To investigate this important but underexplored issue, we introduce the first benchmark for indirect prompt injection …

abstract applications arxiv attacks benchmarking copilot cs.ai cs.cl embed however integration language language models large language large language models llms microsoft microsoft copilot prompt prompt injection prompt injection attacks risk type