SESSION II: AI + Cyber: AI Powered Cybersecurity (INTERMEDIATE)
Fridays, October 16 – November 20, 1:30 PM – 5:30 PM ET
Description
This professional development course provides faculty with practical strategies for integrating artificial intelligence into cybersecurity instruction within the next 12 months. Participants will examine how AI is changing cybersecurity work, explore the risks associated with AI-enabled systems, and evaluate tools that can enhance student learning while keeping cybersecurity and technology programs aligned with evolving threats and workforce demands.
Participants will work with local large language models using Ollama, retrieval-augmented generation and research platforms, AI-assisted cybersecurity tools, and secure coding assistants. These technologies will be used alongside established cybersecurity frameworks, defensive tools, vulnerability-assessment methods, and instructional design practices.
The course is organized around a continuing scenario involving a fictional technical college responding to AI-related instructional and cybersecurity challenges. Across five intensive Friday afternoon sessions, participants will complete guided demonstrations, controlled cyber-range labs, instructor-provided scenario injects, curriculum-development activities, and peer reviews. Each participant will redesign an existing instructional activity and develop a practical 12-month implementation roadmap for their home institution.
The course addresses three connected areas of practice:
- Securing AI applications, models, data, retrieval pipelines, and tool integrations
- Using AI to support cybersecurity research, analysis, defense, and secure development
- Recognizing and responding to adversaries who use AI to increase the speed, scale, or effectiveness of cyberattacks
Commercial platforms may be demonstrated, but the course emphasizes transferable capabilities rather than dependence on a particular vendor. Core activities will include local, open, preserved-output, or institutionally approved alternatives whenever possible.
NOTE: This track is a repeat from Summer 2026 in Ohio. Participants who previously completed this course are not eligible to register for this track again.
Certification Prep
N/A.
Objectives
- Analyze emerging AI technologies, including local language models, retrieval-augmented generation frameworks, research platforms, coding assistants, and agentic cybersecurity tools, to determine their relevance, risks, and potential application within existing academic programs.
- Evaluate cybersecurity instructional tools, digital platforms, and lab environments using criteria such as learning alignment, technical accuracy, privacy, security, accessibility, cost, licensing, hardware requirements, and workforce relevance.
- Use technical evidence and authoritative sources to validate AI-generated claims, identify hallucinations or unsafe recommendations, and communicate appropriate levels of confidence and uncertainty.
- Design or revise classroom activities, lesson plans, labs, assessments, or project-based learning experiences that integrate AI with cybersecurity tools and industry practices while preserving student accountability and measurable skill development.
- Develop practical syllabus guardrails addressing acceptable AI use, disclosure, attribution, academic integrity, data protection, and required evidence of student work.
- Implement a realistic 12-month action plan for introducing AI-enabled cybersecurity tools, resources, and labs at the participant’s home institution.
Pre-requisites
- Be active faculty members or instructors teaching cybersecurity, information technology, computer science, or a related technology discipline.
- Have the ability to modify or influence course content within the next 12 months.
- Bring an existing course, lesson, lab, assignment, or instructional area that can be revised during the course.
- Possess basic digital literacy and familiarity with common instructional technologies and learning management systems.
- Have a foundational understanding of cybersecurity concepts such as networking, system security, application security, risk management, vulnerability assessment, or security operations.
- Be comfortable reviewing basic terminal output, logs, scan results, or source code.
- Demonstrate a willingness to evaluate new technologies critically and translate course resources into practical student-learning experiences.
- Advanced programming, machine-learning development, and previous experience running local language models are not required.
Required Textbook
None.
Suggested/optional Textbook
None. Current standards, frameworks, advisories, documentation, and instructor-provided resources will be used instead of a fixed textbook.
At-Home Computer Requirements
- Laptop or desktop computer running Windows, macOS, or Linux
- Modern AMD Ryzen 5, Intel Core i5, Apple Silicon processor, or equivalent
- Minimum 16 GB RAM
- Minimum 100 GB available disk space
- Reliable broadband internet connection
- Administrative permission to install approved software
- Hardware-virtualization support when using the local virtual appliance
- GPU optional
The primary virtual appliance will target supported x86-64 virtualization environments. Participants using Apple Silicon or another ARM64 system will receive an architecture-appropriate appliance, container-based alternative, or hosted environment designed to provide equivalent lab outcomes.
If you do not have access to a machine that can run a VM in VirtualBox, reach out to the instructor to ensure you have access to a cloud-provisioned environment.
Participants using institution-managed computers should verify software-installation, virtualization, and network-access permissions before the course begins.
Please note that content is subject to change or modification based on the unique needs of the track participants in attendance.
Agenda
Oct 16: Local AI, Privacy, and Trust
- Participants deconstruct common AI myths and examine what generative AI can and cannot reliably accomplish in cybersecurity. They use Ollama or an equivalent local platform to compare local and hosted execution, examine privacy and data-handling implications, and practice the CO-STAR prompt-engineering framework.
- In the first scenario mission, participants must determine whether Northbridge Technical College should permit faculty to process student or institutional information with an AI system. The accompanying lab requires participants to classify the data, select an appropriate processing environment, construct a structured prompt, and validate the resulting output.
Oct 30: Curriculum Injection and Academic Integrity
- Participants complete a hands-on curriculum-injection sprint using an existing lesson, lab, assignment, or course module from their own institution. They examine how generative AI may allow students to bypass intended learning and redesign the activity around authentic performance, decision-making, process evidence, technical validation, and student accountability.
- The Northbridge scenario introduces a legacy cybersecurity assignment that can be completed by AI without demonstrating the intended competency. Participants revise the assignment and draft practical syllabus guardrails covering acceptable AI use, attribution, disclosure, prohibited data, evidence retention, accessibility, and academic integrity.
Nov 6: AI-Assisted Research, RAG, and Threat Intelligence
- Participants compare open-web AI research with source-bounded analysis. Perplexity AI may be used to demonstrate rapid web discovery, while NotebookLM or an approved equivalent is used to analyze a curated collection of standards, advisories, threat reports, and institutional evidence.
- The lab introduces conflicting sources, circular citations, unsupported claims, and potentially malicious instructions embedded within source material. Participants create a claim ledger, verify consequential citations, identify possible prompt injection, and produce a confidence-qualified threat assessment using authoritative resources such as NIST, MITRE ATT&CK, MITRE ATLAS, CISA, and relevant industry reporting.
Nov 13: AI-Assisted Cyber Operations and Authorized Security Assessment
- Participants examine how AI co-pilots can support—but should not replace—human interpretation of vulnerability scans, terminal output, network evidence, and security telemetry. Activities emphasize scope control, false-positive analysis, authorization, human approval, and verification of AI-recommended actions.
- Within the isolated AI + Cyber Range, participants investigate a Northbridge system using approved scan results and lab evidence. Instructor-controlled injects introduce misleading banners, conflicting findings, incomplete telemetry, and unsafe AI recommendations. Participants then redesign a legacy penetration-testing or vulnerability-assessment lab so that students are evaluated on evidence, interpretation, decision-making, and responsible tool use rather than command execution alone.
Nov 20: Secure Code Review and Institutional Implementation
- Participants use Claude AI, another approved coding assistant, or preserved coding-agent output to review a vulnerable Northbridge application. They compare AI findings with static-analysis results, distinguish confirmed defects from false positives, evaluate proposed patches, identify incomplete fixes, and develop appropriate regression tests.
- The course concludes with participants finalizing and peer-reviewing their 12-month implementation roadmaps. Each roadmap identifies proposed curriculum changes, required resources, institutional stakeholders, policy or approval needs, faculty-development requirements, anticipated risks, implementation milestones, and measures of success.
Instructor

Dr. Frazier Smith brings a unique combination of technical expertise and educational leadership to the field of information technology and artificial intelligence education. His research centers on the application of artificial intelligence in career development programs and evaluating the effectiveness of post-secondary career and technical education programs in preparing students for industry.
With over a decade of experience spanning corporate technical training and higher education, including leadership roles at VMware and Broadcom, Dr. Smith has developed expertise in curriculum design, workforce development, and educational technology. His technical foundation spans cloud infrastructure (AWS, Azure, Google Cloud), virtualization technologies, cybersecurity, and network architecture, complemented by hands-on experience in data center design and systems administration.
As co-Principal Investigator on an NSF grant focused on “Generating Artificial Intelligence Talent” and a key contributor to developing one of the first associate-degree AI programs in North Carolina, he works at the crossroads of AI innovation and workforce readiness. His doctoral research and practical experience position him to address critical questions about how post-secondary institutions can best prepare students for careers in diverse technology fields.