UMERCS BLOG

Matt Hosburgh Matt Hosburgh

Cyber Deterrence and the New Era of Active Defense

The focal point can no longer be on just your organization’s defenses. Maybe Bruce Lee said it best: “Don't think. FEEL. It's like a finger pointing at the moon. Do not concentrate on the finger, or you will miss all of the heavenly glory.”

The cyber world is not the same as the physical world. I think we all mostly agree with that (unless you feel like you’re in the real-life Matrix). But there are kinetic, or physical, implications should an attacker manipulate a system that controls some sort of industrial process. Further, attackers who wish to do harm to critical infrastructure can, in some cases, invoke damage or impact safety. Two immediate examples come to mind. If you haven’t watched the video of the Aurora Generator test conducted by the Idaho National Labs, it provides a proof-of-concept of this reality. Secondly, the recent discovery of the TRISIS malware reveals that the Safety Instrumented Systems (SIS) of certain Industrial Control Systems (ICS) are at risk. What would the results be if this were to be successful? The long game is yet to be seen; however, it is a very real possibility that a cyber attack could be leveraged in conjunction with a physical attack to maximize the effectiveness.

Active Defense and Why Offense is Necessary

Active Defense (AD) is hotly contested and often brings mixed emotions. Part of this debate stems from an inconsistent definition. The DoD defines AD as “the employment of limited offensive action and counterattacks to deny a contested area or position to the enemy.” In the Tallinn Manual, which is the Internal Law Applicable to Cyber Warfare, Active Cyber Defense is defined as: “A proactive measure for detecting or obtaining information as to a cyber intrusion, cyber attack, or impending cyber operation, or for determining the origin of an operation that involves launching a pre-emptive, preventive, or cyber-counter operation against the source.” Others still maintain that attacking back cannot be a part of AD, or even in terms of traditional defense.

Why not?

What is meant by attacking?

What if we called it interacting?

A few years back, Christopher Hoff gave his keynote on the topic. In his talk, he made a reference to Jiu Jitsu and mixed martial arts. The example looked at two practitioners grappling. From innocent bystander’s perspective, it was difficult to see who was actually on offense and who was defending, but one was attacking and the other was actively defending. In Jeet Kune Do (JKD), the martial art heavily influenced by Bruce Lee, there are concepts that helped to guide action. For example, the stop hit, which is a method of preemptively striking before the attacker strikes you—but not before knowing they are going to strike you.

In terms of the digital realm, AD is an incredible concept that can help to perform the digital stop-hit.

TO BE CLEAR: attacking doesn’t necessarily mean flinging exploits at a loosely formed target.

The book Offensive Countermeasures: The Art of Active Defense spells out the varying degrees of interaction with an attacker. Dubbed AAA, the continuum is Annoyance, Attribution, and Attack. Within these As, are techniques that can be leveraged to preemptively defend against an adversary. Truly, it’s a mindset, and one we should shift our thinking towards. As a threat hunter, one of the most beneficial stages to look for adversaries is the Lateral Movement stage. This has become a normal and accepted practice. But why? Why are we okay with an active adversary moving around our networks? What if we could actively deploy defenses that helped to alert earlier, more granularly, and provided means to interact, dare I say, attack back? In military defense, or even in some Somali pirate cases, a unit or ship in the defense would take some form of action should an enemy enter into particular proximity. In some cases, if the enemy fired their weapons, authorization was granted to fire back. Fighting fire with fire isn’t the goal and doesn’t always translate well in the cyber realm, but to the point that defense CAN and often does include forms of attacking back is valid.

Physical vs Cyber Deterrence

In a recent blog by Schneier on Security, Bruce calls out an example surrounding 2016 presidential election where cyber deterrence was taken into consideration. The US was cautious to retaliate in the cyber realm due to estimated or perceived cyber capability of Russia.

It brings to mind physical deterrence. A house guarded with a security system, fences, and perhaps dogs might look less appealing to rob than one without. The problem, as seen from the attacker’s perspective, is one of detection. More specifically, if the attacker thinks they will be caught (or harmed in the action), the chances are less that they will launch the attack. They may look to a more appealing target or another avenue.

Deterrence and Active Defense (Annoyance)

Enter the first A of Active Defense: Annoyance. Picture an attacker in the early reconnaissance phase of an attack. MITRE has the whole PRE-ATT&CK matrix (soon to be consolidated into one ATT&CK Matrix) that looks at the varying attacker techniques BEFORE they get into an organization. This is the space, if proactively engaged with, can help to fend off an attack before it is even launched—remember the stop hit analogy? What if during the OSINT gather step, the attacker discovered a URI for the organization that was “legit?” They then, perhaps, start to spider this URI for further directories or files of interest. Unbeknownst to the attacker, one directory is actually a trap for the spider. On access, and only discovered by an attacker, the page generates random data to frustrate the crawler. Gone un-monitored, the spider would run until manually stopped. From a detection standpoint, any interaction with this particular resource would generate a high-fidelity alert. Done at scale, AD becomes more than just a way to detect, but to deter an adversary before they are able to get a foothold.

Conclusion

Inconsistent definitions create confusion and mixed emotions. Peeling back the red-tape or emotions can reveal what we’re all after—better security. The point here is not to determine if attacking back is a viable security technique. Rather, the point is that we have been tireless fighting a losing battle in security. The mindset must change for this to change. This will take work. It won’t be easy, but I believe, if done right, we can engage our adversaries in a more neutral space, rather than from within our organizations. The focal point can no longer be on just your organization’s defenses. Maybe Bruce Lee said it best: “Don't think. FEEL. It's like a finger pointing at the moon. Do not concentrate on the finger, or you will miss all of the heavenly glory.”

Read More