LLM Agents can Autonomously Hack Websites

[Prat011]

Summary

In recent years, large language models (LLMs) have become increasingly capable and can now interact with tools (i.e., call functions), read documents, and recursively call themselves. As a result, these LLMs can now function autonomously as agents. With the rise in capabilities of these agents, recent work has speculated on how LLM agents would affect cybersecurity. However, not much is known about the offensive capabilities of LLM agents.

Details

In this work, it is shown that LLM agents can autonomously hack websites, performing tasks as complex as blind database schema extraction and SQL injections without human feedback. Importantly, the agent does not need to know the vulnerability beforehand. This capability is uniquely enabled by frontier models that are highly capable of tool use and leveraging extended context. Namely, it is shown that GPT-4 is capable of such hacks, but existing open-source models are not. Finally, it is shown that GPT-4 is capable of autonomously finding vulnerabilities in websites in the wild. These findings raise questions about the widespread deployment of LLMs.

Implementation Guidance

• Investigate the methods used by GPT-4 to autonomously hack websites.
• Explore the limitations of existing open-source models in performing similar tasks.
• Develop safeguards to prevent misuse of LLM agents in cybersecurity.

Reference

LLM Agents can Autonomously Hack Websites

Tags

• LLM
• Cybersecurity
• Autonomous Agents
• GPT-4