Using Weaponized Machine Learning in Cyber Offensive Operations

Using Machine Learning in cyber defensive operations has proved to be highly efficient especially for fast pattern detection. There is yet an application in using Machine Learning in cyber offensive operations in order to improve existing skill set of human operators, or enable large scale offensive operations, otherwise hard to do, without extensive manpower. Machine Learning can be a solution to the complexity of present-day world structure, it can support full autonomous work mode operations and can support asymmetric operations by being the perfect invisible enemy. Having a fully autonomous system that can launch multiple attacks in multiple domains (social, infrastructure, telecoms) simultaneously, can be valuable in a world of interconnected networks. For military operations, it’s obvious that in the era of 4^th generation warfare, such solutions might give an advantage over the other combatants. Furthermore, Machine Learning can become a formidable weapon if used right in the era of 5^th generation conflicts, where it can target the individual itself, escalating effects to groups of people. The paper presents a generic framework in using Machine Learning in offensive cyber operations as a solution to the present-day expansion of cyber operations on foreign and national territory.