TATAR Lab Securing AI. Quantifying Cyber Risk.
A research lab at the University at Albany, SUNY, working on the security and safety of AI systems, the economics of cyber risk, and the protection of critical infrastructure.
Our mission
Advancing security through research
We combine risk analysis, economics, engineering, and policy to answer questions no single discipline can settle on its own.
Measure what is actually risky
We build benchmarks, datasets, and evaluation frameworks that turn vague claims about AI and cyber risk into numbers that can be compared and audited.
Follow risk through systems
Risk rarely stays where it starts. We model how it propagates across model supply chains, insurance portfolios, and interconnected infrastructure.
Translate research into decisions
Our work is aimed at the people who act on it: security teams, insurers, regulators, and boards. We put findings in the language each of them uses.
What we work on
Research focus areas
Six threads run through the lab's current and past work. They overlap by design.
AI Security & Autonomous Agents
We study what AI agents can actually do in offensive and defensive security, measuring how far they get on realistic tasks and what closes the gap between a cheap model and a frontier one.
LLM Risk, Safety & Evaluation
We develop structured ways to score the security posture of language models, working from established taxonomies and tracing how risk in a base model propagates to everything derived from it.
Cyber Insurance & Risk Transfer
We study how to price cyber risk, how catastrophic exposures behave, and where policies written before AI leave coverage ambiguous.
Economics of Cybersecurity
Security budgets are set under uncertainty and the incentives rarely point where the risk is. We apply risk analysis, operations research, and economics to investment and risk mitigation prioritization.
Critical Infrastructure
Attacks on infrastructure do not stop at the system they hit. We model how disruptions cascade through interconnected sectors, including healthcare, industrial control, space, and maritime.
AI Incidents & Governance
Understanding how AI systems fail in the world requires a reliable record of those failures. We work on collecting that record and on evaluating the tooling meant to govern it.
At scale
Multi-institutional Initiatives
Beyond individual projects, the lab helps lead two multi institution centers.
VICOR
Virtual Institute of Cyber Operations and Research
A Department of Defense funded cyber institute run with Florida International University. VICOR builds the operational skills of future cyber leaders through applied training, competitions, internships, and research placements.
CATIS
Cyber and Terrorism Insurance Studies
A national, NSF supported center on catastrophic cyber and terrorism risk, anchored at the University at Albany with the University of Michigan. CATIS spans risk modeling, insurance and capital markets, security controls, systemic AI risk, and policy.
Where the work lands
Research with a constituency
Government & defense
Work supported by and conducted with NSF, the Department of Defense, the Air Force Research Laboratory, the Office of Naval Research, NSA, NIST, and NATO.
Industry & insurance
Collaboration with insurers, actuarial bodies, and risk practitioners on cyber risk quantification, third party risk, and the coverage questions raised by AI.
Students & workforce
Curricula, cyber exercises, competitions, and funded research placements that move students into real operational problems and into the cybersecurity workforce.
Open research output
Benchmarks, datasets, evaluation harnesses, and agent tooling released publicly so results can be reproduced, contested, and built on.
Who leads it
Team
Dr. Unal Tatar
Associate Professor in the Department of Cybersecurity at the University at Albany’s College of Emergency Preparedness, Homeland Security and Cybersecurity. Before joining UAlbany he led Turkey’s national Computer Emergency Response Team and advised NATO’s Centre of Excellence for Defence Against Terrorism. He is the lead principal investigator of the DoD funded VICOR institute and directs the NSF supported CATIS center, with grants and contracts, and has published two edited volumes and peer reviewed papers in venues including Risk Analysis and Computers & Security.
Graduate Students
- Hakan T. Otal
- Brianna L. Bace
- Divij Jain
Undergraduate Students
- Shayaan Khaliq
- Tyler H. Merves
- Joseph M. Escobar
- Michael H. Conaway
Non-Resident Researchers
- C. Cagri Akgunduz
- Thora Knight
Contact
Work with the lab
We are glad to hear from prospective graduate and undergraduate students, and from collaborators in government, industry, and insurance. A short note about what you are working on is the best place to start.
Get in touch- utatar@albany.edu
- Visit
- ETEC Building, Room 350P
1220 Washington Avenue
Albany, NY 12222 - Code & data
- github.com/TATAR-LAB