Dynamic spectrum sharing (DSS) is a crucial solution to the ever-increasing wireless spectrum shortage. DSS systems typically use an automated frequency coordinator called a Spectrum Access System (SAS), which uses both dedicated and crowdsourced spectrum sensors to enable on-demand, cost-effective spectrum sensing with extraordinary geographic coverage and granularity. Crowdsourcing-assisted DSS systems have long faced security and privacy concerns. The project’s novelties are the development, concept prototying, and evaluation of TrustSAS, an innovative AI/ML-powered trustworthy SAS that utilizes both dedicated and crowdsourced spectrum sensors to ensure comprehensive and efficient spectrum utilization while maintaining strong security and privacy. The project's broader significance and importance are the exploration of the latest advances in AI/ML technologies to enable trustworthy DSS and the contribution to the advancement of scientific research in AI/ML-driven wireless and mobile security systems. <br/><br/>The proposed research includes three integrated thrusts. Thrust 1 develops VeriSS, a pioneer framework for verifiable spectrum sensing that employs graphical neural networks to analyze spectrum-sensing data and evaluate their trustworthiness even in the presence of potentially compromised spectrum sensors. Thrust 2 investigates TrustSML, a novel framework that builds upon VeriSS to achieve fast, accurate, and trustworthy localization of multiple simultaneous spectrum misusers. Thrust 3 studies PriCCS, a novel spatial crowdsourcing framework that enables the TrustSAS system operator to perform cost-effective and accurate spectrum-sensing task assignment while providing strong location privacy guarantees to participating crowd workers. The project's deliverables, including tutorials, talks, publications, and open-source toolkits, will be made publicly available online. The proposed research will be integrated into curriculum development, enhancing undergraduate research experiences and fostering broad participation in computing.<br/><br/>This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.