Project Grant 2622944

The National Science Foundation Division of Graduate Education awarded $500,000 to the University of Arizona on August 15, 2026, to develop an educational platform for training cybersecurity professionals in adversarial malware defense and AI-enabled cyber security.

The deliverables consist of two integrated components: an Adversarial Malware Generation and Defense (AMGD) curriculum and an AMGD workbench. The curriculum will define a body of knowledge spanning adversarial machine learning, malware analytics, and AI-enabled cyber defense, informed by the NICE Cybersecurity Workforce Framework and industry feedback from Rapid7. It will include modular learning units covering adversarial machine learning, malware detection and analysis, generative AI, ethical hacking, and robust system design. The AMGD workbench will provide a safe, interactive environment where students can implement, benchmark, and analyze white-box and black-box adversarial malware attacks and defenses using Python notebooks, large language models, reinforcement learning, and genetic programming. The project will employ iterative evaluation and refinement of both components through formative and summative assessment.

The award is issued under the STEM Education program (CFDA 47.076). Work is performed in Tucson, Arizona, with a period of performance extending through July 31, 2029. The University of Arizona operates as the Arizona Board of Regents for administrative purposes.