Towards Authorship Attribution of AI-Powered Obfuscated Malware

  • Wang, Zhaohui (Principal Investigator/Chief Investigator A)

Project: HDR ProjectMasters by Research

Project Details

Description

As malware authors leveraging advances in artificial intelligence (AI) to automatically generate multiple offspring of obfuscated malware binaries, the tasks of reverse engineering and attributing authors have become increasingly complex. Nevertheless, most research on malware authorship attribution lacks both realism and diversity, often failing to incorporate the style of the malware author or group. Therefore, one primary challenge lies in the creation of a dataset that consists of obfuscated binaries linked to known author. With our proposed dataset, we can benchmark the effectiveness of malware authorship attribution models against AI-powered metamorphic malware.
StatusNot started

Fingerprint

Explore the research topics touched on by this project. These labels are generated based on the underlying awards/grants. Together they form a unique fingerprint.