EvilModel: Hiding Malware Inside of Neural Network Models

2021/07/19 by Zhi Wang, Chaoge Liu, Xiang Cui · 3 voices · 4 citations
Computer Science · #Advanced Malware Detection Techniques #Adversarial Robustness in Machine Learning #Network Security and Intrusion Detection

paper · pdf · doi:10.1109/iscc53001.2021.9631425

Abstract

Delivering malware covertly and evasively is critical to advanced malware campaigns. In this paper, we present a new method to covertly and evasively deliver malware through a neural network model. Neural network models are poorly explainable and have a good generalization ability. By embedding malware in neurons, the malware can be delivered covertly, with minor or no impact on the performance of neural network. Meanwhile, because the structure of the neural network model remains unchanged, it can pass the security scan of anti-virus engines. Experiments show that 36.9MB of malware can be embedded in a 178MB-AlexNet model within 1% accuracy loss, and no suspicion is raised by anti-virus engines in VirusTotal, which verifies the feasibility of this method. With the widespread application of artificial intelligence, utilizing neural networks for attacks becomes a forwarding trend. We hope this work can provide a reference scenario for the defense on neural network-assisted attacks.

Cited by

• NeuPerm: Disrupting Malware Hidden in Neural Network Parameters by Leveraging Permutation Symmetry
• SASER: Stego attacks on open-source LLMs
• On the (In)Security of Loading Machine Learning Models
• The Art of Hide and Seek: Making Pickle-Based Model Supply Chain Poisoning Stealthy Again

Discussions

• EvilModel: Hiding Malware Inside of Neural Network Models [pdf] [hn, 85 points, 28 comments]
• EvilModel: Hiding Malware Inside of Neural Network Models [lobsters, 8 points, 0 comments]
• Your favourite LLM may be hiding malware! An interesting research paper presenting "EvilModel" shows how attackers can embed full binary payloads of malware in any standard LLM for malicious attacks: [bsky, 1 points, 0 comments]

Related

• Enhancing Robustness of Deep Neural Networks Against Adversarial Malware Samples: Principles, Framework, and AICS'2019 Challenge
• Adversarial Malware Binaries: Evading Deep Learning for Malware Detection in Executables
• A Survey of Trojan Attacks and Defenses to Deep Neural Networks
• Do You Trust Your Model? Emerging Malware Threats in the Deep Learning Ecosystem
• Hacking Neural Networks: A Short Introduction