Malicious Code

Malicious code research focuses on understanding and mitigating the threats posed by harmful software, encompassing its creation, detection, and impact. Current research emphasizes the vulnerabilities of large language models (LLMs) to generating or being manipulated into producing malicious code, alongside the development of novel detection methods using transformer-based architectures and graph-theoretic algorithms for identifying malicious actors in distributed learning environments. These efforts are crucial for enhancing software security, protecting users from cyberattacks, and improving the robustness of AI systems themselves.