"Applying AI Techniques in Cybersecurity, Counterterrorism, and International Security"
Artificial Intelligence (AI) has shown promise as a valuable tool for protecting against malicious actors. AI has been used to help predict terrorist attacks, destabilize terrorist networks, and mitigate cyberattacks in real-time. A newly established Northwestern University lab will assist in developing and deploying AI technologies that will serve as solutions to global threats. The new Northwestern Security and AI Lab (NSAIL) is conducting AI research relevant to cybersecurity, counterterrorism, and international security issues. V.S. Subrahmanian, Walter P. Murphy Professor of Computer Science in Northwestern Engineering and a faculty fellow at the Northwestern Roberta Buffett Institute for Global Affairs, along with his collaborators, recently presented a number of NSAIL projects aimed at addressing issues at the ever-changing intersection of AI and security. The Northwestern Cyber Early Warning System (NCEWS) is a decision model system being developed by the lab to manage known and zero-day vulnerabilities. NCEWS, which builds and improves on a previous system developed by Subrahmanian, mines ongoing discussions about a given vulnerability using an ensemble of predictors and a combination of natural language methods and social network analysis. NSAIL also developed a model called DIPS (Detected, Infected, Susceptible, and Patched) to predict how badly a network will be affected by a new piece of malware through the use of a human health model for cybersecurity. The team looked at false alarm rates to see what percentage of alarms raised by security products are legitimate, whether the lab can predict which alerts are true, and what percentage of true alarms are missed. While AI plays a significant role in cyber defense, Subrahmanian expressed concern about how hackers will use AI to learn how defenses are used and predict whether a defender will detect a planned attack. He also discussed how hackers could use AI to create previously unknown types of malware, attack graphs, and deepfake-enabled phishing vectors. This article continues to discuss recently highlighted NSAIL projects that apply AI in cybersecurity, counterterrorism, and international security.