Yigitcan Kaya
Name Pronunciation (Turkish): Yee-it-JAHN
In Spring 2027, I will join the Department of Computer Science at Indiana University Bloomington’s Luddy School as an assistant professor. Previously, I was a postdoctoral fellow at UCSB’s SecLab, working with Giovanni Vigna and Chris Kruegel, supported by the Intelligence Community (IC) Postdoctoral Research Fellowship, which I received for my proposal on building AI-driven antifragile cyber defenses. I earned my PhD in Computer Science from the University of Maryland, College Park, advised by Tudor Dumitraş.
I work at the intersection of AI and security: I build AI that makes security work more effective, and I uncover where deployed AI creates new attack surfaces. My recent work develops self-improving security agents (24), uses LLMs to discover zero-day vulnerabilities (23), traces web attacks to their root causes at scale (22), and exposes prompt-injection risks in widely deployed chatbots (16). Earlier, I identified overthinking in adaptive neural networks and showed how attackers can exploit it to erase their efficiency gains (3, 7, 11), demonstrated the fragility of deep networks to single-bit hardware faults (4), and developed practical threat models for adversarial machine learning (1, 2).
My research has received an Amazon Research Award and has been featured in VentureBeat and MIT Technology Review.
I have mentored 20+ undergraduate interns and junior researchers in trustworthy AI (through NSF REU, Summer at MC2, and the ACTION Institute), contributing to multiple top-venue publications (7, 9, 13, 16) and seven PhD placements in leading programs.