Sunday, May 9, 2010

Military Grade Malware (Part 1)

Not all malware is created equal. Of the 50k-80k new and unique malware samples received daily by the mainstream anti-virus companies, there's a lot of scope for variety. Most of the samples are merely serial variants being pumped out as part of a barrage of criminal campaigns, and then there's a sizable handful of custom crafted malware that (for the most part) is generally unsophisticated botherware and spyware, but occasionally you'll uncover a few very crafty and sophisticated malware samples mixed in there.

In a lot of cases, these particularly sophisticated malware samples only manage to get caught up in the wash of general malware samples because of some circuitous and "unlucky" compromise paths - or because they're several months old and the "discoverers" have finished reaping the reward of having investigated them. Most of the really interesting bespoke malware samples rarely come via mainstream discovery and sample sharing systems though - in fact the majority of them rarely go beyond the virtual walls of the organization or government department that were targeted or victimized by them.

Given all the discussions about Advanced Persistent Threats (APT), Advanced Malware and Next Generation Malware (NG Malware), I thought it was about time to disclose some of the techniques being used within the commercial world in the production of such sophisticated malware... hence this blog entry being the first in a series covering "Military Grade Malware".

Military Grade Malware
I use the term "Military Grade Malware" to encompass the following key concepts:
  1. A legal contractual agreement exists between the professional software development team and the purchasing organization.
  2. The expectation is that the "product" will be used for purposes beyond financial and criminal fraud.
  3. The intended distribution of the malware will be limited in scope and typically only be deployed in very specific environments.
  4. The malware is designed to be stealthy and continue to operate for extended periods of time - typically against a sophisticated adversary.
Why are these important? The vast majority of malware circulating around the Internet and infecting both home and corporate systems are clearly designed for criminal purposes. More often than not, they're heavily weighted towards data theft and financial fraud. While the authors of the malware may or may not be criminals themselves (e.g. many of the popular DIY construction kits are sold commercially, by licensed companies, as "Remote Administration" tools) - they are designed to operate on popular operating systems and commodity hardware.

In the past I've used the term "weaponized" to encompass malware that makes use of exploit material as part of its critical operations - but this term only extends so far.

Exploit Weaponization
There are plenty of boutique security consulting organizations out there that offer "weaponization" services. They will typically review and study the latest vulnerability disclosures, develop reliable exploits for use against specific operating systems (e.g. an exploit for a popular Vietnamese instant messaging client running on Microsoft Windows XP SP3 with the Vietnamese language pack installed), and pass the final QA-checked exploit on to their client.

Most of the organizations I've come across that provide this kind of service have strong affiliations with their local government. That said though, a handful of them are more mercenary and will provide their weaponized exploits to other "friendly" governments. I'll point out at this stage though that this is a wholly different arrangement compared to vulnerability research teams working within companies that develop commercial vulnerability scanning and exploitation tools.

The provisioning of (reliable) weaponized exploits will generally be governed by formal legal contracts. It's not easy work though. Many people see the plethora of public vulnerability disclosures and hear about the odd zero-day exploit doing the rounds, but the development of reliable exploits that meet the contractual demands of the client is not a simple task. A company that can deliver a half-dozen ruggedly reliable weaponized exploits each year is doing very well - and will be compensated accordingly.

Malware Weaponization
The weaponization of malware in my opinion generally only encompasses the binding of a "standard" malware component to a particularly good/reliable/weaponized exploit.

For example, a client may have a preferred Remote Access Trojan (RAT). This RAT is consequently bound to the latest weaponized exploit - i.e. the RAT is merely the payload of the successful exploitation.

In another example, a versatile malware agent may support a library of exploits that it can use to worm and propagate around a targeted network. In this case, the weaponized exploit is constructed to be compatible with the malware agent and is added as an "update".

Both examples would fulfill the generic term "weaponized malware", but there is a difference between this type of malware and what I'd tend to term "Military Grade" malware, since military grade malware may or may not actually make use of weaponized exploit materials.

What are the features and techniques of military grade malware? I'll begin to cover those details in subsequent blog posts...

No comments:

Post a Comment