AI Research and Education

PNW faculty and students have been active in applied AI research in cybersecurity.

AI-empowered PLC-based CPS Security: automates the security vulnerability assessment for Programmable Logic Controller (PLC)- based cyber-physical systems (CPS). Machine learning and generative AI technologies are being used to:

1. generate super AI agents that can construct a “CPS Digital Twin” for the CPS system based on a PLC traffic analysis tool developed by the PI’s research team;
2. automate the generation of security assessments and mitigations using RAG (Retrieval-Augmented Generation) Technology;
3. quality assurance and trustworthiness of RAG;
4. least privilege assurance of AI agents for the PLC-based CPS.

Other applied AI research directions include:

1. LLM-enabled UAV systems for autonomous aerial navigation, decision-making, and domain-specific task execution,
2. Industrial ML Models and Time Series of GPTs,
3. Quantum machine learning for IoT and CPS systems,
4. AI and machine learning technologies for gun detection,
5. Security and safety of deep learning models, and
6. Natural language processing methods for multimodal AI and building AI systems integrating diverse modalities for advanced reasoning and trustworthy decision-making in industrial AI.

• Alger, J. Rogers, A. and Tu, M. LLM-Generated Countermeasures for IoT Cyberattacks. Journal of Military Cyber Affairs.  August 2026.
• Alger, J. a. (2025). Anomaly Detection of Network Layer Attacks against Cyber Physical Systems Using Machine Learning and Deep Learning Techniques. Journal of Military Cyber Affairs.
• Calix, R. (2023). A Dataset of CFD Simulated Industrial Furnace Images for Conditional Automatic Generation with GANs., (p. TMS 2024 153rd Annual Meeting & Exhibition Supplemental Proceedings ).
• Calix, R. (2024). Machine Learning-Based Regression Models for Ironmaking Blast Furnace Automation. MDPI Dynamics.
• Calix, R. (2025). A Preferences Corpus and Annotation Scheme for Human-Guided Alignment of Time-Series GPTs. MDPI.
• Dai, D. (2024). Dai, W., Singh, Y.P. and Zhang, R., 2024, December. Multi-agent Simulation for Mass School Shootings. 2024 IEEE International Conference on Big Data (BigData 2024) (pp. 2714-2723). IEEE.
• Dai, W. (2024). Analyzing Mass School Shootings in the United States from 1999 to 2024 with Game Theory, Probability Analysis, and Machine Learning.
• Guntupalli, K, Tu, M, and Raja, A. Disrupting LLM-Drone Operations: Roleplay Based Prompting Jailbreak Attack on LLM-Enabled UAV). In Proceedings of the 2026 IEEE 17th Annual Ubiquitous Computing, Electronics & Mobile Communication Conference (UEMCON). September 2026. New York, USA.
• Guntupalli, K. M. (2025). AeroNav: A fine-tuned LLM framework to enhance LLM-UAV’s aerial navigation task performance. 2025 IEEE 16th Annual Information Technology, Electronics and Mobile Communication Conference (IEMCON). IEEE.
• Guntupalli, K. M. (2025). FLY-LLM Sim: A novel integration of UAV and LLM lab platform. 2025 IEEE World AI IoT Congress (AIIoT). IEEE.
• Huang, W. N. (2026). Tell me what to track: Infusing robust language guidance for enhanced referring multi-object tracking. CASSP 2026-2026 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). IEEE.
• Islam, S. (2022). An intelligent privacy preservation scheme for EV charging infrastructure. IEEE Transactions on Industrial .
• Islam, S. (IEEE Intelligent Systems.). A Q-learning Novelty Search Strategy for Evaluating
• Raja, A. (2023). Towards the Security of AI-enabled UAV Anomaly Detection. IEEE International Conference on Communications (ICC). Rome, Italy: IEEE.
• Raja, A. N. (2022). Adversarial Attacks and Defenses Toward AI-Assisted UAV Infrastructure Inspection. EEE Internet of Things Journal.
• Liu, F., Zhang, T., Tao, J. Tu, M. Auditing LLM-Generated Dietary Recommendations: A Reproducible Pipeline for Nutritional and Behavioral Risk Analysis. MDPI Healthcare: Generative AI in Healthcare: Opportunities, Challenges, and Ethical Implications. Major Revision. September. 2026.
• Singh, S. A. (2026). Multivariate Time Series ML Approach for Anomalies Detection in Smart Industry Cyber-Physical System. Proceedings of IEEE Green Technology 2026. Denver: IEEE.
• Srinivasan K, T. A. (2025). Quantum-Enhanced Clustering for Intrusion detection system in IoT devices. 2025 IEEE 22nd International Conference on Mobile Ad-Hoc and Smart Systems (MASS) (pp. 610-615). Chicago: IEEE.
• Stefanek, G. (2024). A comparison of AI models to detect hidden messages in images. Issues in Information Systems, 119-132.

CWCT is an online CyberAI workforce certification-based training program launched in September 2020.  It is a collaborative effort by a coalition of academic institutions and industry partners, including PNW (the leading institution), 8 other NCAE institutions, and other industry partners.

1. CWCT offers five learning paths in cybersecurity and AI. Training courses are offered in 8-week sessions with 45 contact hours per session and have delivered 25 training sessions to the transitioning workforce, primarily veterans and transitioning military.
2. Participants are prepared and supported to earn industry-recognized certifications (CompTIA Linux+ and Security+, EC Council CEH, CHFI, Cisco CyberOps, etc.).
3. CWCT partners with industry and workforce recruiting agencies for professional development and job placement.

Outcome Summary: Through September 2026, offered 400+ training classes with a total of 16,000+ enrollments in 25 sessions to over 4,000 training participants, processed 20,000 applications, supported certifications with exam vouchers, and hosted 18 career fairs and 29 professional development workshops.

This project is supported by National Centers of Academic Excellence in Cybersecurity Grants, housed at the National Security Agency: #H98230-20-1-0351 and #H98230-23-1-0087, as well as the DoD Cyber Academic Engagement Office Grant: HQ0034251E014-210263

• AI-Driven Cybersecurity Solution for Cyber-Physical Systems of Advanced Manufacturing (Faculty Lead: Dr. Michael Tu). PNW Interdisciplinary Research Grant 2024.
• An Agentic AI Framework-Based Modular RAG System for The Automation of CPS System Security Assessment and Control. (Faculty Lead: Dr. Michael Tu). PNW Interdisciplinary Research Grant 2025.
• LLM-enabled UAV system navigation and execution (GenAI, Trustworthy AI, Faculty Lead: Dr. Ashok Raja)
• Industrial ML Models and Time Series of GPTs (GenAI, Transformer Architecture, Faculty Lead: Dr. Ricardo Calix)
• NLP and multimodal AI for advanced reasoning (NLP, LLM, GenAI, Trustworthy AI, Faculty Lead: Dr. Yang Ni)
• AI and ML technologies for gun detection and safety evaluation (ML/DL, GenAI, Faculty Lead: Dr. David Dai)