Using AI for cybersecurity
Help analysts detect, investigate, and respond faster while keeping evidence, escalation, and accountability visible.
Cybersecurity research and engineering
Greenwood Cyber + AI Lab
Cyber Resilience for the AI Era
We build and evaluate defender tools, data-protection patterns, and test suites that make AI systems safer in production.
Scope
Where we focus
The work follows three risk paths: how defenders use AI, how sensitive data moves through AI systems, and how AI platforms can be attacked.
Protecting data in AI
Set boundaries for sensitive data across prompts, retrieval, applications, logs, and model outputs.
Protecting AI platforms
Find and reduce ways AI systems can be attacked, misused, monitored poorly, or trusted too soon.
Principles
How we work
The lab uses AI to support accountable security work and human judgment without letting automation own the outcome.
Human Accountability
The lab uses AI to support accountable security work and human judgment without letting automation own the outcome.
Data Boundaries
Sensitive data boundaries are defined before deployment, including prompts, retrieval, logs, and outputs.
Evidence-Driven
Recommendations need tests, measurements, or enough detail for others to evaluate and reuse the work.
Automation Limits
Automation earns scope gradually, with review points, limits, and a path for human intervention.
Resilience
Systems should fail safely, recover well, and remain observable under pressure.
Featured Work
Project examples
These examples show the kind of work the lab supports: understand the system, focus the risk, test what matters, and turn findings into practical decisions.
Location
Based in Greenwood
The lab is in a region shaped by energy, aerospace, defense, finance, higher education, and critical infrastructure, making Greenwood a practical base for work on cyber resilience and AI security.
Leadership
Meet the team
The lab’s leaders bring security, engineering, and field experience to projects shaped with customers and partners.
Michael Salazar
General Manager
Joseph Aguayo
Technical Director
Contact
Work with the lab
The lab partners with researchers, defenders, educators, students, workforce groups, community organizations, and security teams on practical AI security problems.
Microsoft Forms
Send a request
The form collects your name, email address, and a short description so the lab can review and respond to your request.
Do not include passwords, sensitive personal data, customer confidential information, or details of an active security incident.
Open Microsoft FormsMicrosoft will use the information you provide to review and respond to your request. See the Microsoft Privacy Statement.