AegisShield: Democratizing Cyber Threat Modeling with Generative AI

The increasing sophistication of technology systems makes traditional threat modeling difficult to implement and scale. This is especially true for small organizations that lack resources and expertise. This research develops and evaluates AegisShield, a generative AI-enhanced threat modeling tool that implements frameworks such as STRIDE and MITRE ATT&CK to automate threat model generation and provide systematic assessments. By integrating real-time threat intelligence from sources including the National Vulnerability Database and AlienVault’s Open Threat Exchange, AegisShield produces streamlined, accurate, and accessible threat descriptions. Our assessment of 243 threats from 15 case studies and over 8,000 AI-generated threats shows that AegisShield significantly reduces complexity (p < 0.001), produces outputs that are semantically aligned with expert-developed threats (p < 0.05), and achieves a statistically validated 85.4% success rate in mapping threats to MITRE ATT&CK techniques (p < 0.001). Simplifying threat modeling through automation and standardization helps under-resourced organizations get ahead of risks. As a result, this promotes a wider adoption of secure-by-design principles and encourages a more secure ecosystem. Note: This is the public dissemination copy of the doctoral praxis submitted to The George Washington University in partial fulfillment of the requirements for the degree of Doctor of Engineering. The official archival version, including accreditation and front matter, is available through ProQuest and GW ScholarSpace.