Strahinja (Straho) Janjusevich, a native of the small Balkan country of Montenegro, says his life has unfolded in an unexpected way and he is deeply grateful. After high school, he was selected to represent the United States and studied cyber operations and computer science at the Naval Academy in Annapolis, Maryland. Since then, I have continued researching cybersecurity and am currently a second-year master’s student in the Technology and Policy Program (TPP) sponsored by the MIT Institute for Data and Systems Society (IDSS). His research with the MIT Institute for Information and Decision Systems (LIDS) and the MIT Maritime Consortium team aims to use artificial intelligence to improve the cybersecurity of critical maritime infrastructure, considering both the technology and the solution’s policy framework.
“My current research focuses on applying AI techniques to cybersecurity issues and investigating the policy implications of these advances, particularly in the context of maritime cybersecurity,” says Janjusevich. “Representing my country at the highest levels of education and industry has given me a unique perspective on cybersecurity challenges.”
Janjusevich’s path from Montenegro to Maryland was created by a program that allowed selected students from allied countries to attend the U.S. Naval Academy. Janjusevic graduated with a dual bachelor’s degree in Cyber Operations and Computer Science. His undergraduate experience provided him with the opportunity to work with the U.S. military and National Security Agency, exposed him to high-level cybersecurity operations, and sparked his interest in tackling complex cybersecurity challenges. During my undergraduate studies, I also interned at Microsoft developing cloud incident response tools and at NASA visualizing solar data for research applications.
After graduating, I realized that I needed more knowledge, especially in the areas of AI and cybersecurity. TPP immediately appealed to him because of its dual focus on rigorous engineering innovation and the policy analysis needed to deploy it effectively. Janjusevich’s experience at TPP was a big change from his time at the U.S. Naval Academy, with a different pace and environment. He particularly appreciates being able to expand his understanding of different fields of study and apply the discipline and knowledge he gained during his time at the Academy.
“My experience with TPP has been great,” says Janjusevich. “This group is really small, so it feels like a family, and everyone is working on diverse, high-impact issues.”
Reduce risk of new technology
Janjusevic’s thesis brings together the fields of cybersecurity, AI and deep learning, control theory and physics, with a focus on the security of maritime cyber-physical systems, particularly large legacy vessels. If the networks of these vessels were hacked, it could not only have severe economic consequences, but also cause significant damage to national security.
“Strajo is working to outwit maritime GPS spoofing,” says Saurabh Amin, Edmund K. Turner Professor of Civil Engineering. “Such attacks have already thrown ships off course in disputed waters. His approach overlays physically-based wake models with deep learning to capture threats that cannot be detected by any single method alone. His expertise has been invaluable in advancing our research in threat modeling and attack detection.”
This research uses advanced threat modeling and ship dynamics to train an AI system to distinguish between legitimate maneuvers and spoof signals. This includes building a framework that uses an internal LSTM (long short-term memory) autoencoder to analyze signal integrity, while using physically-based forecasting capabilities to predict vessel movement based on environmental factors such as wind and sea conditions. By comparing these predictions to reported GPS locations, the system can effectively distinguish between natural sensor noise and malicious spoofing attacks. This hybrid framework is designed to empower, rather than replace, human operators, providing verified navigation data that allows observers to distinguish between technical glitches and strategic cyberattacks.
Sanjusevich was able to use his industry experience to enhance his academic research. In the summer of 2025, he interned with the network detection team at Vectra AI, an AI cybersecurity company. There, he investigated the potential threats that new technologies could pose, specifically AI agents and the emerging standard for AI agent communication, Model Context Protocol (MCP). His research demonstrated how this technology can be reused for autonomous hacking operations and advanced command and control. This research on agent AI security risks was recently published in the preprint “Hiding in AI Traffic: Exploiting MCP for Agent Red Teaming with LLM.”
“We gained practical insight and hands-on experience on how data science teams can use AI models to detect anomalies in networks,” says Janjuusevic. “This work within the industry directly influenced the anomaly detection models in my research.”
International policy perspective
“Mr. Strajo not only brings a high level of intellect and energy to research on commercial maritime cyber-physical security, but also strong instincts from his naval training that resonate deeply with the research effort and ground it in actionable policy,” said Fotini Cristia, Ford International Professor of Social Sciences, director of the IDSS and MIT Center for Socio-Technical Systems Research, and leader of the MIT Maritime Consortium.
Janjusevic participates in the Maritime Consortium’s cybersecurity efforts, a collaboration between academia, industry, and regulators focused on developing technology solutions, industry standards, and policy. The consortium includes collaboration with several international members, including Singapore and South Korea.
“Policy elements are very important in AI cybersecurity because the field is changing very quickly and the consequences of hacking can be very dangerous,” says Janjuusevic. “I think there is still a lot of need for policy work in this area.”
Janjusevich is also currently helping organize two major upcoming conferences. One is the Harvard European Conference in February, which brings together government officials and diplomats from around the world, and the other is the Technology and National Security Council in April, a collaboration between Harvard and MIT that brings together top leaders from government, industry, and academia to address critical national security challenges.
“I strive to find a position where I can impact and advance the cybersecurity field with AI, while also leading collaboration and innovation between the United States and Montenegro,” Janjusevich said. “My goal is to be a bridge between Europe and the United States in the areas of national security, AI, and cybersecurity, and to provide my knowledge to both sides.”
