Synthetic Cancer -- Augmenting Worms with LLMs

TL;DR

This paper introduces a novel malware that uses large language models to automatically rewrite code for evasion and socially engineer email replies, demonstrating new cybersecurity risks posed by LLMs.

Contribution

It presents a minimal prototype of LLM-based malware that can evade detection and socially engineer recipients, highlighting emerging cybersecurity threats.

Findings

LLM-based malware can effectively evade signature detection.

The malware can socially engineer email replies to spread itself.

Demonstrates the need for advanced cybersecurity measures against LLM threats.

Abstract

With increasingly sophisticated large language models (LLMs), the potential for abuse rises drastically. As a submission to the Swiss AI Safety Prize, we present a novel type of metamorphic malware leveraging LLMs for two key processes. First, LLMs are used for automatic code rewriting to evade signature-based detection by antimalware programs. The malware then spreads its copies via email by utilizing an LLM to socially engineer email replies to encourage recipients to execute the attached malware. Our submission includes a functional minimal prototype, highlighting the risks that LLMs pose for cybersecurity and underscoring the need for further research into intelligent malware.