Published:
December 20 2010, 01:23 PM
|
1 Comment(s)
by
Sumner Blount
I happened upon an interesting blog recently, "WikiLeaks: Stronger Access Mgt. Needed," by Eric Chabrow that considered the role of ineffective access management in the WikiLeaks debacle. Merritt Maxim has also posted some interesting blogs on this topic, including, "We have seen the Security Enemy and it is Us."
Chabrow's blog also references an interesting study by the Ponemon Institute, sponsored by CA Technologies, "Security in the Trenches." One of the more interesting findings of this study was the people on the "front lines of security" tend to see higher risk of security breach (such as information loss) than their superiors did. These IT security folks understand the challenges of protecting key applications, systems, and information, and tend to have a more nuanced and often pessimistic view of the overall security of their infrastructure than the upper level executives do. The lesson of this study is that all levels of the organization (from HR to IT and Senior Mgmt) need to be involved in an honest evaluation of the effectiveness of any given set of IT security controls. And if audits uncover control deficiencies, organizations must be committed to fixing these as expeditiously as possible.
But, getting back to WikiLeaks....We could debate for days the true causes of this security breach. Clearly, there was no effective access governance system in place that would prevent this person from gaining access to such a huge number and wide variety of confidential documents. In addition, there were no controls over what could actually be done with the documents once they were obtained. An effective data loss prevention system could probably have prevented most or all of those documents from being emailed inappropriately to someone outside the military. An automated and effective procedure for analyzing access event log files, or file transfer logs, might have been able to either recognize the problem or highlight an area for further manual analysis. In short, there is plenty of blame to go around in terms of non-existent or ineffective controls on access to information and its use within this particular military system.
This experience is a very painful national lesson as to the requirements for strong security and control across not only identities and access, but also information use. First, there must be up-to-date, widely communicated, and attested-to information use policies. Users must know instinctively what they can and cannot do with the information that they access. Granted, in the WikiLeaks case this would not have prevented this breach. Still, in many organizations information use violations occur simply due to lack of knowledge or understanding on what the policies proscribe. Next, there should be technology in place to prevent or detect certain policy violations. In most cases, users cannot be trusted in all cases to abide strictly to your information use policies, whatever they may be. Just because someone needs access to a given document or collection of information, that does not ensure he or she won't use that information in a way that violates policy and is damaging to an organization. Finally, there should be periodic or continuous auditing of access event logs to look for suspicious activity that might precede a breach. This will not prevent it in all cases, but could detect it soon enough to minimize the impact.