Skip to content

Prompt Injection Attacks on Language Models

Authors: Huang et al. Year: 2023 ArXiv/Link: https://arxiv.org/abs/2301.10226

Summary

Taxonomy and analysis of prompt injection attacks where user input can override system prompts.

Key Concepts

  • Prompt injection
  • Attack taxonomy
  • Input validation
  • Model robustness
  • Security vulnerabilities

Impact

Identified critical vulnerability class in LLM systems

Category

Jailbreaks & Attacks