Re:In Cryptography, Advances in Program Obfuscation
by nitrogen on Thursday January 30, 2014 @08:52PM
(#7154769)
If the malware depends on a sufficient number of system details to decrypt itself, then no amount of code analysis will help decrypt the inner layers. As you say, it's like obtaining a password, and a sufficiently strong password (or sufficiently unique set of system details) cannot even be brute forced before the heat death of the universe.
[ Parent
| on HN ]
Re:In Cryptography, Advances in Program Obfuscation
by jjoonathan on Friday January 31, 2014 @12:36AM
(#7155473)
If by "no amount of code analysis will help" you mean to exclude techniques such as running the damn thing on a known-vulnerable box/VM, then sure. But very few reverse engineers operate under that restriction.
It's like DRM: the user must have the key, so the only security you can have lies in obfuscation.
Re:In Cryptography, Advances in Program Obfuscation
by jlgaddis on Friday January 31, 2014 @12:54AM
(#7155527)
> ... such as running the damn thing on a known-vulnerable box/VM, ...
But in the case of Flame, wasn't that (one of) the big problem(s)? That the researchers didn't know what, exactly, comprised a known-vulnerable host?