While external hacking incidents have been making the headlines recently, we can't ignore the insider threat.
I'm fortunate to live in Seattle, where I have access to many professional security organizations. One organization is the Cloud Security Alliance, which holds a meeting each month where security professionals gather to learn about new technology and threats. Last month, the speaker was one of the authors of Verizon's 2014 Data Breach Investigations Report. Each year, Verizon analyzes more data as an increasing number of organizations provide data from their breach experiences. As was expected, there was a lot of data from breaches and incidents from external sources—that is, from sources not originating from within the organization breached. While that data was the topic of several interesting discussions, a different topic—centered around the internal threat (breaches and incidents originating from within the organization)—also provoked some interesting discussions. This month's article focuses on the internal threat and what the data does—and doesn't—show. Here's what I learned.
One can make the assumption that the insider threat is less of an issue than the external threat. And it's an obvious assumption given that the breaches and incidents documented as external are far greater than those documented as internal. But as our speaker pointed out, organizations are under no obligation to report insider breaches or incidents as long as they don't fall under breach notification laws or violate a state or federal law. Here's an example: It's doubtful that any organization would ever report that someone who had excessive authorities was able to accidentally upload a modified spreadsheet and change a production database file, making it unavailable until the database was restored. Yet that classifies as an incident. Therefore, it goes without saying that the number of internal breaches and incidents is significantly higher than what is reported in the survey. The question no one can answer is: how much higher? Regardless, the insider threat should not be ignored.
So let's dive a little deeper.
First, let's define what's meant by a breach or an incident. Here are the definitions used in the Verizon report:
- Incident: A security event that compromises the integrity, confidentiality, or availability of an information asset.
- Breach: An incident that results in the disclosure or potential exposure of data.
- Data disclosure: A breach for which it was confirmed that data was actually disclosed (not just exposed) to an unauthorized party.
Not surprising, by far the largest form of abuse is the misuse of privileges—think Edward Snowden and his abuse of privileges at the NSA. But unlike Edward Snowden, who was a system administrator, system administrators are the smallest percentage of users that abuse their privileges. Most abusers are cashiers (in the retail sector) and end users. Yes. End users are a larger threat than system administrators. What does this mean to the IBM i community? Users should only have the capabilities (i.e., special authorities and group profile assignments) that they need to perform their job functions. User profiles are often copied based on the profile of an existing user. Often, that user has more authority than is required or is an experienced user with more job responsibilities than the new employee. But rather than give only the authorities needed by the new user, the profile of the existing user is blindly copied and the new user now has more privileges than necessary to perform their job.
Tip #1: Create user profiles with only the privileges required to perform their job duties.
How are the privileges misused? Usually from within the four walls of the enterprise itself, meaning that people didn't surreptitiously log on from home or a remote location to use their own or stolen credentials; they did it right from their desk using a LAN-connected workstation.
Tip #2: Provide a safe (anonymous) method for employees to report suspicious behavior or changes in other employees' behavior.
Tip #3: Educate employees to lock their workstations when leaving their desks and have a short time-out (e.g., five minutes) on your network that enforces this policy to prevent someone from using (and abusing) someone else's credentials.
What are users stealing? Credit cards, bank account numbers, and social security or social insurance numbers are probably the first types of data that come to mind. And while that's true, other types of data are also being stolen: trade secrets, healthcare information, and general personnel information. Trade secrets are being stolen at an increasing rate and most often are taken within 30 days of an employee resigning from the company. Healthcare information is being sold on the black market and used to obtain prescription medications illegally. But general personnel information…why is that being stolen? All data is typically being sold for the individual's (that is, the person doing the stealing) financial gain.
The scary part of that equation is that there's evidence of a growing influence of organized crime being part of the actions related to data theft. While there's no hard evidence in the report, the discussions during the presentation led me to believe that the general personnel information could be sold and later used as an entry point into the organization. For example, every organization has a naming convention for email addresses. With the general personnel file, the group buying this information now has the entire organization's email address list and can start spear phishing to gain further access—that is, make very targeted phishing attacks. You know, those emails that try to get you to click on some link. A spear phishing attack may target specific individuals and often are made to look like they originated from a high-level company executive. The hackers' goal is to convince someone to click on a link and unwittingly download malicious code that will work its way into the network, thereby providing a launch point for future attacks and exploits.
Tip #4: Consider your employee list as highly confidential.
Tip #5: Consider revoking users' access to confidential data when they announce that they are leaving the organization. Or walk them to the door straight away and revoke all network access immediately. (Don't forget to notify your cloud providers to revoke cloud application access as well.)
Tip #6: Delete inactive IBM i user profiles to ensure that former employees or contractors do not have access to your IBM i systems. (A statistic from Intermedia and Osterman Research shows that the majority (89%) of respondents to their survey retained access to at least one application from a former employer after leaving.)
Why am I writing about the insider threat? Because I'm shocked by the number of IT managers and system administrators who continue to assert, "Our employees don't know how to do that" or "Our employees would never to do sort of thing"—meaning access data inappropriately. That thinking must stop or the data residing on our IBM i systems will continue to be vulnerable to misuse. Here are three steps you can take:
- Stop thinking in terms of the skills and motivation of the people who work for your organization and start thinking about whether or not your organization's data is vulnerable.
- Determine what data stored on your IBM i system might be the most valuable to sell or be meaningful to a group outside of your organization.
- Analyze who has access to this data and take steps to secure it (that is, restrict access) if it's accessible to more users than require it to perform their job duties.