Shir Bernstein is a Masters student and AI & Cybersecurity Researcher at Ben-Gurion University. Her research focuses on the security vulnerabilities and biases within Large Language Models. Notably, her M. Sc. thesis on exploiting LLM pattern bias was accepted to the top-tier NDSS 2026 security conference.
Outside of her research, Shir has a pragmatic and humorous approach to technology, enjoying code jokes and even engineering solutions for personal annoyances like keyboard language errors. She actively engages with the cybersecurity community by attending events such as Microsofts BlueHat IL conference.
Unique fact: Her research demonstrates how to hijack static code analysis in advanced AI models by exploiting their inherent pattern biases.
Read the full overview →Behavioral traits and communication patterns that shape how this person evaluates opportunities, builds trust, and makes decisions in professional settings.
Dominance, influence, steadiness, and calculativeness scores with guidance on how this person prefers to communicate and decide.
Discover additional public profiles from our index.
Looking for someone else? Search here for anyone.