The Remote Metamorphic Engine: Detecting, Evading, Attacking the AI and Reverse Engineering

Presented at DEF CON 24 (2016), Aug. 5, 2016, 3 p.m. (60 minutes)

As a matter of fact, it is all about time to reverse engineer the most complex piece of code. Code complicity techniques are usually used just to increase the time and effort needed for reverse engineering. The desired effect of code complicity can be magnified using mechanisms that decrease and narrow the allowed time frame for any reverse engineering attempt into few milliseconds. Such approach can be applied using a metamorphic engine that is aware of the time dimension.

Beyond metamorphic applications for AV evasion, in this talk, we will present a novel approach to resist and evade reverse engineering using a remote metamorphic engine that generates diversified morphed machine code of a very short expiration lifetime. Our approach is based on a client-server model using challenge-response communication protocol made of morphed machine code rather than data. We will show how any reverse engineering attempt on such model will be forced to execute or emulate the morphed code. Thus the code will always have an upper hand to detect, evade and attack the reverse engineering environment. Our approach is immune to static code analysis as the functionalities and the communication protocol used are dynamically diversified remotely and do not exist in packed executable files. On the other hand, clock synchronized morphed machine code driven by a remote metamorphic engine would trap dynamic RE attempts in the maze of metamorphism. One that is immune to code tampering and reversing by detecting the non-self.

We will present the fundamental difference between metamorphic and polymorphic techniques used to evade AV compared to the ones that can be used to resist RE. We will show how a remote diversified metamorphic self-modifying code with a very short expiration lifetime can detect, evade, and resist any code analysis, reverse engineering, machine learning and tampering attempts.


Presenters:

Links:

Similar Presentations: