From Prompt Injections to SQL Injection Attacks: How Protected is Your LLM-Integrated Web Application?

Rodrigo Pedro, Daniel Castro, Paolo Molina, Nuno Santos

Abstract

Demonstrates how prompt injection can be chained with traditional web attacks (SQL injection, XSS) in LLM-integrated applications.

Categories

Tags

Framework Mappings

Cite This Resource

@article{llmsec202400045,
  title = {From Prompt Injections to SQL Injection Attacks: How Protected is Your LLM-Integrated Web Application?},
  author = {Rodrigo Pedro and Daniel Castro and Paolo Molina and Nuno Santos},
  year = {2024},
  journal = {USENIX Security 2024},
  url = {https://arxiv.org/abs/2308.01990},
}