I know the Ponemon Institute has a sponsor for each of their studies but the recently released Second Annual Cost of Cyber Crime Study does contain some really valid findings that CSO´s should take into account.
Their study highlights just how many companies are immature in the detective and reactive controls following a breach. I would suggest that for data theft and fraud breaches, most of the deployed controls as so ineffective that often, the occurance of a breach is detected though a source outside of their organisations. A great example of this is the TJX breach where the incident was reported by TJX officials around a month after an extensive fraud had occurred.
There are many reasons for this including but not limited to zero day hacks, or ineffective intrusion detection systems. However, as Ponemon points out in their findings, “companies using SIEM were better able to quickly detect and contain cyber crimes than those companies not using SIEM”… Yes, the sponsor or this study is a SIEM provider. More importantly though, it does point to the fact that audit log information is a key source of information in the detection capabilities. However as recognised year after year by another study known as the Deloitte Global Financial Services Security Survey; within the top 5 internal/ external audit findings is “Audit trails/ logging issues”. I.e., organisations are still wrestling with collection, analysis and protection of audit log data. CSO´s need to place greater emphasis on using some of their rapidly reducing IT budgets on log collection, analysis and protection tools. Because as Ponemon points out “Cyber attacks can get costly if not resolved quickly…the average time to resolve a cyber attack by a participating organisation was 18 days at an approximate average cost of $415k with malicious insider attacks taking more than 45 days”… I know of more than a few log management solutions that cost less than that.
OK, now lets assume that you have a solution to detect and fix a breach but now want to prosecute or even have to defend a prosecution. Most companies are missing a critical capability that will cause such a significant pain point in litigation scenarios. Specifically most organisations have not deployed capabilities so that the electronic data to be digital evidence ready. I will write more on this in an upcoming post, however if you cannot prove the electronically stored data´s authenticity, it will not be usable as evidence. A great source to identify the controls you need to consider are located in BS 10008 – Evidential weight and legal admissibility of electronically stored information.
In addition to the digital evidence point, it is now a known fact that the hacker of today is not out to be noticed. In fact they have always preferred to stay undetected until they want to get noticed. To do this they delete or modify data that may show their activity. In most hacking 101 books/ papers, there are sections on how to conduct a stealth attack and how to remain undetected by deleting or modifying log data. It’s obvious that if the log data has been modified the time to detect and respond to an attack will be substantially increased.
A professor at one of the first lectures that I had in Information Security asked the question: “What is the most dangerous thing a hacker can do?” After going through a list of responses by the class, he suggested; “It is to stay hidden and seep corruption into the organisations digital data that it even gets into the backups and yield an unrecoverable digital data environment…. This type of effect can put a company out of business”. Now, years on, I recognise that this is not the worst a hacker can do, i.e. they can cause an outage of a power grid, mayhem at a nuclear plant etc. However with respect to normal business it does make me wonder how many organisations were subjected to this type of attack when the UK´s Ministry of Defence published the warning “Foreign hackers ‘putting UK firms out of business’“.
Author: Nadeem Bukhari