Security Attacks on LLM-based Code Completion Tools

TL;DR

This paper reveals significant security vulnerabilities in LLM-based code completion tools, demonstrating successful attacks that compromise proprietary data and expose sensitive user information, highlighting urgent need for improved security measures.

Contribution

It introduces targeted attack methodologies specific to LCCTs, exposing vulnerabilities and demonstrating effective jailbreaking and data extraction attacks on popular tools like GitHub Copilot.

Findings

99.4% success rate in jailbreaking GitHub Copilot

46.3% success rate in attacking Amazon Q

Extraction of sensitive user data including emails and addresses

Abstract

The rapid development of large language models (LLMs) has significantly advanced code completion capabilities, giving rise to a new generation of LLM-based Code Completion Tools (LCCTs). Unlike general-purpose LLMs, these tools possess unique workflows, integrating multiple information sources as input and prioritizing code suggestions over natural language interaction, which introduces distinct security challenges. Additionally, LCCTs often rely on proprietary code datasets for training, raising concerns about the potential exposure of sensitive data. This paper exploits these distinct characteristics of LCCTs to develop targeted attack methodologies on two critical security risks: jailbreaking and training data extraction attacks. Our experimental results expose significant vulnerabilities within LCCTs, including a 99.4% success rate in jailbreaking attacks on GitHub Copilot and a 46.3% success rate on Amazon Q. Furthermore, We successfully extracted sensitive user data from GitHub Copilot, including 54 real email addresses and 314 physical addresses associated with GitHub usernames. Our study also demonstrates that these code-based attack methods are effective against general-purpose LLMs, such as the GPT series, highlighting a broader security misalignment in the handling of code by modern LLMs. These findings underscore critical security challenges associated with LCCTs and suggest essential directions for strengthening their security frameworks. The example code and attack samples from our research are provided at https://github.com/Sensente/Security-Attacks-on-LCCTs.