"aI Teaches Itself": Exploring Young Learners' Perspectives on Artificial Intelligence for Instrument Development

Abstract

Children encounter and use artificial intelligence (AI) with regularity, but the depth of their understanding of AI is often limited. In service of growing an AI and technology-literate K-12 population, it is important for young learners to engage in AI learning activities early and often. To foster the design of AI curricula, it is essential to understand what young children already know and how they feel about AI. The nascent field of AI-related self-report instrument development focuses largely on adult populations or AI's use in specific contexts, such as medicine. There remains a critical need to develop an AI attitudinal survey for young learners (ages 9 to 11). Building upon the extant survey development work of those in education and AI, we have designed a brief survey on students' self-efficacy for AI, interest and motivation toward AI, and attitudes toward AI. We used cognitive interviewing processes to ensure the items in the survey were readable and understandable by young students. Preliminary findings indicate young students have mixed understanding of what AI is, what it can do, and how they feel about AI. We discuss implications for researchers and practitioners and provide an overview of our continuing efforts to validate this instrument.