Cyber attacks are becoming increasingly more common. In fact, a recent report by Forbes predicts that the global cost of cyber crime will reach $2 trillion by 2020. But it’s not just citizens who are being targeted; government organizations and institutions are also being targeted by cyber attacks.
In an effort to protect against such attacks, the Defense Advanced Research Projects Agency (DARPA) has issued a proposal to develop software for automatically detecting and neutralizing large-scale attacks from compromised devices. It’s not uncommon for nefarious individuals to gain access to users’ devices, typically through common tactics like phishing and virus deployment. Once the individual has access to these devices, he or she can use them to attack other systems and networks. This collective group of compromised devices is called a botnet, and that’s exactly what DARPA wants to protect against.
Known as the “Harnessing Autonomy for Countering Cyber-adversary Systems” project, it will explore new systems and technologies to protect against cyber attacks carried out by botnets. Of course, this isn’t the first time that DARPA has expressed interest in cybersecurity. Just last year, DARPA hosted the Cyber Grand Challenge, during which white-hat coders and security experts were asked to build cybersecurity systems that could protect against attacks.
Under the “Harnessing Autonomy for Countering Cyber-adversary Systems,” DARPA will award contracts to various businesses to develop new cybersecurity measures. The biggest problem associated with botnet attacks is that fact that they compromised devices often go unnoticed by the respective owner. When someone’s computer is taken over, for instance, he or she may not realize; therefore, the hacker who controls the computer may continue to use it unbeknownst to the rightful owner. DARPA’s new project will explore new ways to stop these devices automatically, without the need for the owner to manually shut them down.
“The goal of the Harnessing Autonomy for Countering Cyberadversary Systems program is to develop technologies for accurately identifying malicious cyberadversary infiltrated networks, generating software exploits for large numbers of known (n-day) vulnerabilities, and creating safe, reliable, and non-disruptive autonomous software agents that can be inserted in the compromised networks via the n-day exploits to safely and reliably neutralize cyberadversary software agents,” wrote DARPA.
DARPA further explained that one possible solution is to develop an autonomous AI system that works in a similar way as an actual human. The system could theoretically detect botnet attacks, after which it neutralized them.