Hiding Malware inside a model of a neural network
Security Affairs
JULY 26, 2021
Tests conducted by the experts demonstrated how to embed 36.9MB of malware into a 178MB-AlexNet model within 1% accuracy loss, this means that the threat is completely transparent to antivirus engines. 58 antivirus engines were involved in the detection works, and no suspicious was detected. ” states the paper.
Let's personalize your content