Insider threat detection is one of a number of recent developments in cybersecurity. The problem of insider threat has become so great that the US government set up the National Insider Threat Task Force (ITTF) in September 2011. The Task Force’s main remit is to establish an insider threat detection and prevention policy for government agencies. You need to ensure that your organization is taking similar steps to prevent insiders from sabotaging your system or stealing sensitive data.
What is an insider threat?
Everyone can understand the concept of a threat to a system. Most cybersecurity tools aim to block hackers and malware from attacking a network or other company resources. An insider is someone who is authorized to use the company’s system.
While outsider attacks can include many software-based systems, such as remote access Trojans, DDoS attacks, ransomware, spyware, and viruses, insider threats are almost exclusively manual processes. An insider might damage the system by deleting data, removing programs, or altering system configurations but the main focus of an insider threat is data theft.
While damage to or removal of system software can be a problem, there are very few account holders in a business who have the ability to perform these actions in a meaningful way. For example, an employee who uninstalls the productivity suite from his own workstation isn’t really causing much damage to the organization. Corporate-wide applications, such as a web server, or database systems, can only be tampered with by a very limited number of people within the IT department.
Small businesses are more vulnerable to sabotage by their limited number of employees and associates because the smaller software packages utilized by these businesses tend to have fewer distinctions between operators and administrators. Still, even in small businesses, the focus of insider threats is on data issues.
What insiders pose a threat?
There are eight types of employees who pose a threat to an organization:
- Disgruntled employee – A person who feels that she has been unfairly treated by the business, passed over for a promotion, denied overtime pay or expenses, or harbors resentment towards a supervisor or section manager.
- Leaver – A person who has been enticed away by a rival business might be induced to take business data away with him to benefit the new employer.
- Fifth columnist – An employee who has a political bias against the company and its activities and seeks to undermine the profitability or viability of the business.
- Manipulated individual – Someone who is put under social or moral pressure to work against the interests of the business.
- Negligent employee – Someone who doesn’t intend to sabotage the business or disclose information but causes damage through a negligent failure to follow best practices and is likely to be indiscrete with data or system access credentials.
- Overspender – A person who needs more money than she earns and is liable to seek to make money by selling company data.
- Former employee – A person who has left the company but whose system access has not been revoked.
- Whistleblower – A person who seeks to gain moral reward by disclosing a company’s trade secrets.
Minimizing insider threats
Before approaching insider threat detection, an organization needs to implement system controls that make data theft or system damage harder to achieve. Minimizing the potential for employee actions is the first phase of insider threat management.
A strategy for reducing the potential for insider threats falls into two categories:
- Identity and access management (IAM)
- Data loss prevention (DLP)
These two topics are explained in depth below.
Identity and access management
Identity management involves the creation of user accounts that can then be applied to resources. User account management is an effective deterrent to insider threats if it is implemented thoroughly.
Identity management tasks include the creations of levels and types of users. This requires the definition of different types of access required by different business roles. The general user population should fall into one category. This category could be split further by department. So, finance department users and HR staff will need access to different software packages, for example.
By defining the requirements of different job titles, the system administrator can arrive at a set of user profiles. Each profile will be defined by the software that user group needs, the hardware resources they will need to access in order to work, and the data stores that group needs to access. Profiles will be further refined by the level of access to data that each need – read-only, or editing rights.
User profiles that include higher permission levels should be assigned to a very limited number of users. Technician profiles should be highly restricted and those technician profiles that enjoy full superuser privileges should be allocated to only a few members of the organization.
Data loss prevention
Data loss prevention measures are more important for a limited type of data. Companies only store data because it is useful. However, some data types are more important than others. If your industry is subject to legal constraints over data disclosure or industry standards, such as PCI-DSS or HIPAA, you need to focus your highest data loss prevention efforts on the types of data specified in the standards that your company is obliged to follow.
Data loss prevention involves locating all stores of sensitive data and placing extra tracking and protection measures in those locations. As all of the data held by the business is important, all data access events need to be tracked.
Insider threat management
Insider threat management is the process of combining identity and access management with data loss prevention.
There is no system that can give you a 100 percent guarantee that you will never suffer an insider threat. Even if your IT system is the tightest possible, you will still face the threat of pen and paper. As well as effective insider threat IT controls, you will also need to implement physical access controls on your premises, staff profiling, and in-person employee monitoring.
You can refine your insider threat management strategy by implementing transaction monitoring.
User and entity behavior analytics
Automated monitoring systems can easily track every event that occurs in the IT system. However, identifying which event is a security threat is a difficult proposition. Fortunately, a field of competence called user and entity behavior analytics (UEBA) helps with that task.
UEBA is an AI tool and is an advanced type of machine learning. A UEBA system records every transaction that occurs, particularly those that involve access to data. Over time, the UEBA system defines a standard of normal behavior. Once the UEBA system has been operating long enough to make a reasonable estimate of standard behavior, it is easier to spot suspicious behavior.
You can’t issue an instruction to a computer to tell it to look out for suspicious activity because it is almost impossible to define what is and what isn’t “suspicious.” UEBA focuses on working out what regular behavior looks like and then raises an alert when anything that deviates from that standard occurs. This tool enables system security specialists to focus on likely insider threats rather than trying to watch every single transaction.
As no system is 100 percent foolproof it is a good idea to hope for the best, but plan for the worst. You need to put in place systems that will enable you to investigate insider threats once they have occurred. Even if you put very stringent controls in place, someone might just get around them one day.
Logging everything that happens generates a large amount of data. However, storage is very cheap these days so the cost of keeping all of those log messages is minimal. If records of all events are in storage, they can be searched for indicators of an attack. If you ever discover that an insider is trying to steal data or you find out that a data loss event has just occurred. You need to search through logs to look at all of the past activity of that user because it is unlikely that you just happened to catch her at her first attempt.
If you need to show compliance to a data security standard, event logging is a requirement and not an option.
Triage involves focusing your efforts on the most likely source of a threat. UEBA and automated activity tracking help you to identify likely malicious user accounts. Once those candidates are identified, you should apply very close scrutiny to their activities. You can’t sack someone on suspicion of malicious activity; you need proof.
Documents and slide presentations not only contain data but they are a key method of sharing corporate strategies. Electronic files are easy to copy and send out in emails or take out of the building on a memory stick.
The first layer of protection against data theft through documents is to scan all outgoing emails for attachment and block USB sockets on workstations.
Not all email attachments are data theft attempts – it is an accepted business practice to send out quotes, orders, invoices, proposals, sales pitches, and receipts as attachments. Therefore, outgoing attachment scanning needs to be sophisticated.
The further line of defense for document control lies with fingerprinting. If key documents are leaked, it should be easier to trace the insider responsible if all document access and distribution systems mark each copy with an identifier.
Data outflow controls
Under data loss prevention systems, sensitive data is located and tracked with tighter controls. Those controls should pay particular attention to any attempt to send out those data types attached to an email or copy them onto a memory stick. This applies triage to data movement tracking.
All of the above insider threat management strategies rely on activity tracking. Analytical tools need to be applied to event logging systems to join together a series of actions that could contribute towards a data loss event. This is because data theft typically involves several steps. Each of those steps in isolation could seem innocuous. It is only when those events are linked together that the theft becomes apparent.
All of the elements listed above contribute to an effective activity monitoring strategy. Controls on user accounts implemented through an identity and access management strategy limit the accounts that need to be thoroughly monitored. Data loss prevention techniques that identify the stores of sensitive data let activity monitors know which parts of the system to focus on.
So, if a user account with high permission rights accesses a store of sensitive data, you’ve got an activity that needs to be tracked closely. What that user account does next is very important. You don’t want to lock down access to sensitive data completely because that data is stored for reference and without it, many of your company’s departments would cease to function. You don’t want to block those superuser accounts either – they were created for reason.
Activity monitoring uses triage to trip a monitor when access to sensitive data occurs, raise an alarm when a subsequent action falls outside of the normal pattern of working behavior (established by UEBA), and applies scrutiny and investigative procedures that scan through event logs.
Insider threat management tools
You have a lot of data to protect and a lot of activities to watch. Naturally, you can’t do all of that manually. Fortunately, your entire insider threat management strategy can be implemented with just a few specialized tools. Here are two tools that demonstrate how two areas of specialization can cover all of the tasks you need to perform in your insider threat detection system.
These are not the only tools in their fields but we consider them to be the best.
The Security Event Manager from SolarWinds is a category of security software called SIEM. It collects and searches through log messages to look for threats – both intrusion and insider threats.
Log messages are generated constantly and reading them provides a source of live activity data. The Security Event Manager collects log messages from every part of the system and puts them in a common format. This enables the SIEM to search through logs looking for event correlations – patterns of different events that combine to indicate a security breach.
The SEM uses UEBA to establish a baseline of normal behavior against which is can identify anomalies. When a potential insider threat event is detected, the SEM raises an alert, drawing attention to that user account. All log records are stored to enable historical analysis. So, once a user account is identified as displaying suspicious behavior, all activities of that user up to that point can be recalled.
The Security Event Manager is able to implement incident response actions automatically. It can suspend an account, eject a memory stick, block all outgoing emails from a user and log that user off the system. The option to trigger automatic responses is up to the system administrator – an alternative approach would be to let the system raise alerts and then leave the response to a decision-maker once further analysis has been carried out.
SolarWinds Security Event Manager installs on Windows Server and it is available on a 30-day free trial.
Teramind DLP is a data loss prevention system that has templates to adapt its operations according to specific data security standards. This is an excellent tool insider threat protection system for businesses that need to comply to the PCI DSS, HIPAA, ISO 27001, and GDPR standards.
As its name suggests, Teramind DLP focuses on protecting data. It is ideal for blocking intruders and insider threats. The tool starts its service by searching all around the IT system for all data stores. It then searches through every data location for specific types of data and categorizes them. This enables the service to focus on specific data, particularly the types of information that a given data security standard aims to protect.
The search for sensitive data continues throughout the systems service life. All events related to identified sensitive data are tracked and logged. While monitoring data stores, the system also analyzes all user accounts. It implements UEBA to get an overall view of normal activity so it can then identify anomalous behavior. Naturally, unusual activities that touch on sensitive data generates an alert.
In addition to data and user monitoring, the service checks data transfers over the network, in emails, and on memory sticks. Any unauthorized activity gets blocked. The system also includes investigative tools for historical data analysis and response methods plus surveillance systems, such as a keystroke logger that can be applied to suspicious user accounts.
Teramind DLP prevents data loss, monitors for data theft events, and then investigates events after they have happened. This is a complete data protection system for insider threat management.