SBIR Phase I: Agentic Artificial Intelligence System

The broader/ commercial impact of this Small Business Innovation Research (SBIR) Phase I project is in developing an agentic artificial intelligence (AI) powered digital platform. The platform is aligned with human-centered AI principles and adapts to changing contexts while preserving user control and operational reliability. The technical innovation of this project is the development of an agentic artificial intelligence system unlike conventional civic platforms relying on passive notifications, this approach employs bounded autonomous agents operating within transparent, explainable frameworks aligned with human-centered AI principles including fairness, accountability, and robustness. The core technical challenge is designing agents capable of dynamic task sequencing and context-sensitive coordination that respect human autonomy while ensuring reliable performance under uncertainty. The Phase I scope will establish technical feasibility through design and testing of an agentic framework using structured workflows and a controlled participant cohort. The key focus is on designing coordination architecture grounded in responsible AI principles, that integrates interpretable decision logic, behavioral models, and policy-governed workflows to support sustained activities. The system will process contextual information through explainable algorithms, with transparency and accountability as foundational requirements. The methodology includes: (1) constructing secure, privacy-preserving data pipelines; (2) implementing human oversight mechanisms and algorithmic accountability measures; (3) developing decision protocols with built-in auditability; and (4) conducting controlled experiments assessing system stability, engagement persistence, and coordination effectiveness. Natural language generation will provide clear explanations supporting user understanding and informed consent. 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.