Design, Development, and Deployment of Context-Adaptive AI Systems for Enhanced End-User Adoption

TL;DR

The paper addresses the gap between rapid AI/ML capability and end-user adoption by proposing context-adaptive AI systems deployed across disembodied and embodied forms. It outlines a three-stage research program—design, development, deployment—anchored in user-centered methods, including a co-design tool (DeX-AI) and a PL-based, safety-guaranteed robotic AI capable of conflict detection, alternative solutions, and explanations. Key contributions include the DeX-AI design cards for end-user-driven explanations and a PL-methods-based autonomous robot system with automated repair and explainability, along with an explicit plan to integrate RLHF, formal verification, and XAI in future work. The work aims to produce practical tools and guidelines that enhance trust, transparency, and adoption of AI in real-world settings, from workplaces to homes.

Abstract

My research centers on the development of context-adaptive AI systems to improve end-user adoption through the integration of technical methods. I deploy these AI systems across various interaction modalities, including user interfaces and embodied agents like robots, to expand their practical applicability. My research unfolds in three key stages: design, development, and deployment. In the design phase, user-centered approaches were used to understand user experiences with AI systems and create design tools for user participation in crafting AI explanations. In the ongoing development stage, a safety-guaranteed AI system for a robot agent was created to automatically provide adaptive solutions and explanations for unforeseen scenarios. The next steps will involve the implementation and evaluation of context-adaptive AI systems in various interaction forms. I seek to prioritize human needs in technology development, creating AI systems that tangibly benefit end-users in real-world applications and enhance interaction experiences.