What is threat hunting

The premise of threat hunting

It is assumed that there are threats that cannot be automatically detected by the system. This assumes that the threat is already present in the organization's network, undetectable by sensors such as network perimeters and endpoints.
In addition, in advanced attacks such as targeted attacks, it is said that operators (hackers) remotely control infected terminals to achieve their goals, rather than operating malware alone. In other words, even if the organization was temporarily protected by the countermeasures on the defending side, they will consider and implement means to circumvent it. In addition, attackers who have infiltrated the internal network make good use of tools that already exist in the computer and tools that can be used legally to gain control of the internal network, move around, and steal information. They are said to be difficult to find.

threat hunting approach

Threat hunting is an approach that assumes both the defender and attacker situations described above. As mentioned earlier, a fully automated system alone cannot defend against advanced threats in light of the fact that they cannot be discovered automatically. In addition, the attacker is a "person", and the defender considers a bypass method. It is considered
However, manually finding and verifying all possible intrusions is difficult in terms of operational costs. In many cases, machine learning is used to automatically discover patterns that are different from the general public and patterns that have never occurred in the past from data within the organization, and "humans" verify that it is an attack. done in the form

In the endpoint market, there is also the understanding that EDR = threat hunting. However, as mentioned above, it generally refers to the process of discovering threats with the intervention of "people", so please be aware that there are cases where it is used with a different meaning than the original intent. .