ESC

Research

Four connected themes that follow the Lab’s pipeline from vulnerability to threat, evidence, and defense.

Vulnerability Discovery

Finding the weaknesses that make attacks possible, and reporting them through coordinated disclosure.

Active Doctoral Research

Systematic Vulnerability Analysis of ARM-Based Platforms

Developing systematic, AI-assisted methodologies for identifying exploitable vulnerabilities in ARM-based software, with the goal of responsible disclosure through coordinated vendor programs and of informing new defensive models.

Prathamesh Anil Walunj, PhD Student & Research Assistant — Advisor: Dr. Murat Gunestas

Threat Research

Studying how adversaries exploit weaknesses and how they behave once inside.

Manuscript in Preparation

AI-Assisted Threat-Enabled Deceptive Honey Systems

Deceptive systems that lure and profile adversarial behavior while producing realistic, labeled datasets for forensic research.

Advisor: Dr. Murat Gunestas

Digital Forensics

Detecting contradictory and tampered digital evidence, and building investigative methods for new technologies.

Published 2026

Indicator of Inconsistency (IoI) Framework

A novel framework for defining/sharing/detecting contradictory digital artifacts using schema-consistent SPARQL constraints over CASE/UCO ontology-based knowledge graphs. IoIs highlight events of probative value including anti-forensics activity. Published in Forensic Science International: Digital Investigation (2026).

Now an open, community-driven catalog of reusable temporal, structural, and semantic IoI rules, evaluated over 87M+ RDF triples. Investigators and researchers can contribute new anti-forensic scenarios and SPARQL rules.

With former researchers Kismat Kunwar & Raquel Elisabeth Gabriel Pita — Advisor: Dr. Murat Gunestas

Experiments Completed

Forensics of AI-Integrated Systems

Developing forensic methodology for emerging AI infrastructure, where no established investigative practice yet exists.

Advisor: Dr. Murat Gunestas

AI-Assisted Cyber Operations

Applying AI to strengthen vulnerability identification and security operations.

Dataset Completed · Fellowship 2026–27

AI-Assisted Vulnerability Identification in Source Code

Applying machine learning to the identification of security vulnerabilities in real-world source code. A research dataset has been completed, and the project earned a competitive University Graduate Research Fellowship.

Yash Kitey, M.S. Student — Advisor: Dr. Murat Gunestas