It should be upsetting to realize that if an intruder is active in a network, most organizations simply would not know. It’s possible that an outsider gained access to the network and quietly began to explore it and expand their area of control.
In the first few days, the attacker has been able to find and take control of other privileged network credentials using Windows exploits or other tools. Then, over the next few weeks, the attacker used the credentials and gained access to a domain controller that proved to be a pivot point in reaching and taking control of other network segments and servers. Finally, over the next few months, the attacker gained access to all the servers in the data center and began looking for files and data that might have value. Getting to the email server turned out to provide a goldmine of material fit for extortion or other purposes. So many companies conduct the core of their business through email. If confidential emails become public or are provided to competitors, investors or other interested parties, the damage could be catastrophic.
In the scenario I’m relating, the attacker has not yet done anything with the assets that might eventually tip off the company or some third party that a network attack has occurred. For now, the attacker is lingering, choosing the right time to capitalize on their position.
The vast majority of enterprises have no effective means of detecting an active attacker on their network. Is someone lurking in your network? Your company likely has no idea.
Most security is geared toward preventing threats in the first place. Some security tries to detect an initial intrusion; other tries to monitor internal traffic, but is ill equipped to find the operational activities of a targeted external attacker or a malicious insider. Some of these systems may actually find some aspect of the attacker at work, but the indication is likely buried under a flood of hundreds or thousands of other alerts that are mainly false positives. The chance of detecting the attacker and then acting on the alert is infinitesimally small.
Obviously one of the most important capabilities enterprise security should take on is the ability to detect an attacker early in the process before theft or damage can occur. A second capability is also extremely valuable: the ability to know whether the network is free from attackers. We think of this as security assurance.
With the fallout and penalties of a data breach growing increasingly more costly, enterprise executives and boards of directors should start demanding that their security heads provide regular reporting attesting that the network is safe from internal or external attackers. The question to ask should be: is there an active attacker currently on our network? The answer should be definitive and based on full visibility that detects attackers and alerts on their activities, namely the internal reconnaissance and lateral movement necessary to carry out an active attack.
Soon, regulatory bodies will start penalizing organizations that have a data breach if they have not taken available precautions to protect the data. The FTC and SEC both have signaled such a direction, and it seems to be inherent in the General Data Protection Regulation (GDPR) that governs the safe keeping of data on EU citizens. Class action and shareholder litigation will likely both examine this issue and use it to support their claims.
Did the organization employ the means to find an active attacker? Did they continually monitor for the presence of an attacker? Were there any alerts from the security products that warned of suspicious or malicious activities? Did executives and the board hold the security team accountable for detecting, investigating and shutting down attacks? No one is going to want to face an SEC inquiry or TV news cameras being asked uncomfortable questions about why they didn’t have a system and procedures for behavioral attack detection and a process for investigating alerts indicative of attacker activity.
The ability to attest to a network being free from attackers should become a primary metric for security. Today, organizations often report on things like the number of end-user computers hit with malware that required help desk support or the number of vulnerabilities found and addressed in network defenses. These are all fine, but a more meaningful metric would be the sign that the network is safe from a hidden intruder.
Customers and partners may start asking for this kind of certification. Large law firms already must go through significant security reviews with their biggest clients. Companies that take credit cards have to go through PCI reviews. Both types of review processes could greatly benefit from an attestation that the network is attacker-free, particularly since the implication is that if there were an attacker, they would know about it and be able to defeat it. They could set up additional review steps to make sure that all suspicious activity is investigated.
The process of integrating two companies through a merger or acquisition would also benefit from the ability to know that an attacker is not lurking in one of the two networks. In a merger or acquisition, there is considerable IT due diligence, but it does not currently extend to knowing what will happen when two networks are connected. Does a hidden attacker get carte blanche access to everything on the new network?
First, companies need a new level of internal visibility that can precisely detect active attackers. Most enterprises lack this today, largely because they have not even considered it. Many of the tools are quite new, and procedures and strategies have not been developed. It doesn’t mean giving up on preventative security, but it does mean shifting some budget and resources toward network attack detection. And it also means enabling your short-staffed security teams with the right tools to quickly investigate possible attacks once an alert is triggered.
By modernizing their security and changing the mindset that it is important to investigate and triage alerts, enterprises should gain the ability to defeat internally- and externally-based attackers. At the same time, they should start being able to offer security assurance reporting to those accountable for the health and viability of the enterprise.
Sign up for our free weekly e-newsletter for more GRC articles, job postings, GRC events, white papers & more…..click here
Peter Nguyen is Director, Technical Marketing at LightCyber, a leading provider of behavioral attack detection solutions. He is responsible for supporting product evaluations and testing, technical training, technical support for channel partners, management and enhancement of the cyberattack simulation environment and the product demo environment. Peter interacts with customers, prospective customers, channel partners and LightCyber developers and field personnel. He developed the Cyber Attack Training System (CATS) as a way to illustrate how network attacks occur and progress while addressing many of the misperceptions about such attacks.
Peter has been at LightCyber since early 2015. Prior to LightCyber, Peter was responsible for technical marketing for security and emerging products at Juniper Networks and, prior to that, at Proofpoint. He served as Corporate Systems Engineer for Blue Coat Systems while leading the team to develop programs and processes to support field sales engineers. He started at Blue Coat as a Lead Software Quality Assurance Engineer. He began his career at Lightbridge as a Systems Administrator. Peter earned his Bachelor of Science degree in Computer Science and Engineering from the University of California, Los Angeles.