06/04/2026
Study Shows How Large Language Models Undermine Traditional Security Controls
In a recently published paper in the World Journal of Advanced Research and Reviews, Tendai Nemure, a 2026 graduate of the Katz School of Science and Health’s M.S. in Cybersecurity, examines a fast-emerging problem in modern cybersecurity: how large language models (LLMs), the systems behind tools like AI chat assistants, are quietly reshaping how organizations think about security.
“On one side is Zero Trust Architecture, a widely adopted cybersecurity approach built on the idea of never trust, always verify," said Nemure. "On the other is the rapid spread of AI systems being embedded into everyday business operations.”
His research, which followed PRISMA systematic review standards and analyzed 68 sources from major databases and security organizations, identifies four major failure points. These include breakdowns in identity verification, vulnerabilities in memory and context, policy bypasses through autonomous agents and weakened data boundaries in retrieval systems.
Nemure’s paper makes a central contribution: a structured taxonomy that maps how these security gaps occur. The taxonomy connects all seven core principles of Zero Trust Architecture to specific risks introduced by LLMs and defines measurable criteria for how those gaps might be closed.
The findings highlight how AI systems are creating new cybersecurity challenges as organizations rapidly adopt AI tools across everyday operations.
“As organizations rush to deploy AI, innovation often moves faster than security,” said Nemure. “In this new landscape, trust is no longer just about who is acting but about how those actions are formed in the first place.”
Read the full story: https://vist.ly/56n38