Data leakage is the occurrence of an organization inadvertently exposing sensitive information – usually due to a mistake like overlooking a critical vulnerability – to the public internet or unsecure networks. This process increases the chances that data will be taken by malicious actors.
In a worst-case scenario, data “leaks” off of the originating secure network and into the hands of bad actors who will hold the sensitive data for ransom or leak it wider onto more visible platforms and websites.
The National Institute of Standards and Technology (NIST) defines a breach as:
"The loss of control, compromise, unauthorized disclosure, unauthorized acquisition, or any similar occurrence where: a person other than an authorized user accesses or potentially accesses personally identifiable information; or an authorized user accesses personally identifiable information for anything other than the authorized purpose.”
Simply put, a data breach is when data is knowingly accessed in an unauthorized manner. A data leak is when an authorized user mistakenly exposes data to the internet or unauthorized networks, but it technically hasn’t been stolen – yet.
The difference between these two terms is small but important when taking actions to secure the data in question or when reporting on the incident later.
Data leakage occurs as a result of a number of mistakes or oversights – or something that no one in an organization would ever have thought of. Let’s take a look at a few ways data leakage can occur:
The effects of data leaks can be disastrous. But, like with anything in security, so much of the process is about timing. If analysts are able to catch the cause of data leaks early, the overall business may be lucky enough to entirely avoid any negative fallout. Or it may be able to minimize the damage. Or it may have to deal with business- or reputation-altering repercussions.
Waiting until something happens shouldn’t be the priority; it should be planning in case of the event. Damage to reputation is something that can and should be scoped prior to the occurrence of any significant future event. That way, a business and its IT and security organizations will have a playbook to follow in such a situation. This will help minimize lasting negative reputational impact.
Following on from possible large-scale reputational damage, there is a two-pronged effect when it comes to a business’ bottom line: potential ransomware payments to threat actors as well as customers taking their business elsewhere. Businesses could quickly find themselves bankrupt or extinct if they aren’t prepared for the consequences of unintended data leakage.
The amount of time it takes for an organization to return to normal operations will depend on the severity of the security event following a data leak and in-progress initiatives that may have to be fully halted in an “all hands on deck” type data security event. This can cause incredible disruption to a business and create an operational deficit from which it could be near-impossible to return.
The current cybersecurity talent shortage and skills gap only seems to continue to exacerbate as more managed security service providers (MSSPs) are called upon to provide monitoring, detection, and response actions on behalf of clients. Hiring skilled in-house talent can already be a laborious enterprise. Following a breach that causes catastrophic reputational damage? Not likely.
While there are obviously certain data types that are of higher value to threat actors – personally identifiable information (PII), financial- and health-related, etc. – what are some of the main vectors by which data leakage occurs? We’ve covered some of the various functionalities, but let’s now group them by type.
Whether it was initiated by an internal source or perhaps a supply chain partner, to be classified as human error in this sense the act/disclosure/exposure must be unintentional. The root cause of this data exposure or leak might have begun as a misconfiguration during the SDLC and turned into a gaping vulnerability through which high-value data was exposed.
The inciting incident could also be something much less technical. Leaving workstations unattended and accessible while working remotely and lost devices are two such examples of mishaps that occur every day and lead to unintended negative consequences.
For the purposes of this page, we are mainly discussing data leakage in a scenario whereby an internal actor – employee, visitor, contractor, vendor, etc. – would unknowingly leave data unprotected or exposed to potential theft or ransom.
However, if an exposure is leveraged by attackers to more easily steal potentially sensitive data, then this type of leak would be attacker initiated. The responsibility for the exposure, though, still lies with the person or people who were initially tasked with securing the data. But if a door is left open, we can all reasonably assume there aren’t many attackers who wouldn’t throw it wide open and steal sensitive data.
It's entirely possible to effectively prevent sensitive enterprise-level data from being exposed and subsequently leaking onto the public internet or into the data stores of malicious actors.
Whether one of the following preventive options are used as a standalone solution or part of a larger product suite, each organization should keep their unique needs and goals in mind when researching which solution/product is best for their environment.