Key points are not available for this paper at this time.
This doctoral research aims to to design interactive Explainable AI (XAI) tools in response to the challenge of fostering AI literacy among adults without technical expertise. The tools developed thus far focus on edge detection, confidence thresholds, and sensitivity, and were designed based on principles from learning sciences and user-centered design to make AI accessible and ethically reflective. My efforts have resulted in successful design, implementation, and preliminary evaluation. Conducted with 42 adult participants, the study reveals notable improvements in familiarity with and confidence in discussing AI concepts. Qualitative feedback highlights user engagement and enhanced understanding, demonstrating immediate impact and laying a foundation for ongoing work. Looking forward, my work will delve deeper into how non-experts can critically engage with AI's decision-making processes, understand algorithmic trade-offs, and consider how AI can better serve society. This approach broadens AI literacy and leverages cognitive principles for creative and ethical technological interventions.
Maalvika Bhat (Sat,) studied this question.