What type of study is this?

This is a Observational study (also classified as: In Vitro Study).

August 22, 2025Open Access

Automated Malware Source Code Generation via Uncensored LLMs and Adversarial Evasion of Censored Model

RARaúl Acosta-BermejoInstituto Politécnico Nacional JTJosé Alexis Terrazas-ChavezInstituto Politécnico Nacional EAEleazar Aguirre AnayaInstituto Politécnico Nacional

Key Points

Malware generation through uncensored large language models presents cybersecurity risks that require urgent attention.
Evaluation using the pass@k metric reveals the effectiveness of adversarial evasion strategies in creating malware.
This research employs a specialized model fine-tuned to produce malware code in a controlled environment, ensuring safety standards.
Understanding malware from the red team's perspective may lead to improved defense strategies against evolving threats.

Abstract

Malicious programs, commonly called malware, have had a pervasive presence in the world for nearly forty years and have continued to evolve and multiply exponentially. On the other hand, there are multiple research works focused on malware detection with different strategies that seem to work only temporarily, as new attack tactics and techniques quickly emerge. There are increasing proposals to analyze the problem from the attacker’s perspective, as suggested by MITRE ATT&CK. This article presents a proposal that utilizes Large Language Models (LLMs) to generate malware and understand its generation from the perspective of a red team. It demonstrates how to create malware using current models that incorporate censorship, and a specialized model is trained (fine-tuned) to generate code, enabling it to learn how to create malware. Both scenarios are evaluated using the pass@k metric and a controlled execution environment (malware lab) to prevent its spread.

Perguntar à IA

Bookmark

View Full Paper