COUNTERING INNOVATIVE SANDBOX EVASION TECHNIQUES USED BY MALWARE
No ratings
Presented at First 2017 by
Automated behavior-based malware analysis is the core function of security solutions defined as “network sandboxing”. It came to the fore for analyzing and detecting advanced threats over a decade ago. Back then, malware authors had already found ways to evade tools like traditional antivirus, which rely on static analysis, by using techniques such as polymorphism, metamorphism, encryption, obfuscation and anti-reversing protection. Malware analysis sandboxes are now considered the last line of defense against advanced threats. It is important to note, however, that the success of behavior-based malware detection hinges on the behavior exhibited by the file during analysis. If, for some reason, no malicious operations are performed by the file during the analysis, the sandbox concludes that the file under examination is benign. Malware authors are always looking for new, innovative ways to evade sandbox detection by concealing the real behavior of malicious files during analysis. In order to cope with the omnipresent threat posed by malware, we must upgrade our defensive tools to succeed in the ongoing cat-and-mouse game of evasion and detection. We therefore must understand what evasion techniques are successfully employed in the wild. This presentation provides an overview of the state-of-the-art evasion approaches used by malware. We divide these approaches into three categories and explore the various evasion techniques associated with each of these: Evasion by detecting the presence of a sandbox: The first approach uses several techniques to detect the existence of a sandbox. Once a malicious file determines that it is being executed in a sandbox, it alters its behavior in an effort to avoid being detected. Evasion by exploiting weaknesses in the underlying sandbox technology: The second approach directly exploits weaknesses in the underlying sandbox technology or in the surrounding ecosystem. Evasion using time, event or environment based triggers: The third approach exploits the natural shortcomings arising from the fact that sandboxes are automated systems. In an effort to deal with the sheer volume of malware, sandboxes usually only spend a few minutes analyzing each file. By delaying the execution of a malicious payload by a certain amount of time, only becoming active on certain triggers, etc., malware can remain undetected.