Willing but Underequipped: the Attitude- Behavior Gap and Design Requirement for Parent-ai-learner Collaboration
Year: 2026
Type: conference-paper
Study design: Observational Study
Abstract: Artificial intelligence (AI) is increasingly entering children’s home learning, where parents remain central to whether and how it is used. This paper examines parents as willing but under-supported mediators of children’s AI-supported learning. Drawing on a mixed-methods survey of 117 parents of primary-school-aged children attending private schools in China, we measured parents’ attitudes, reported AI use, preferred involvement, and open-response accounts of tutoring challenges, AI-tool difficulties, and desired features. Our findings show a substantial attitude-behavior gap: parents evaluated educational AI favorably but reported lower levels of use. Their preferred role clustered in an active engagement zone, favoring substantial co-engagement rather than non-use or delegation. Desired AI supports aligned with parents’ reported tutoring difficulties, especially explanation, diagnosis of learning needs, engagement, and support for independent thinking. The findings identify design requirements for learner-AI-parent collaboration, suggesting that AI systems for home learning should scaffold both children’s learning and parental mediation rather than functioning only as direct tutors.
Authors: Wang, Jingyun, Shang, Yi, Qi, Xiaofei
Language: English
Subfield: Education
Field: Social Sciences
Domain: Social Sciences
SDG: Quality education
Open Access status: green
OpenAlex ID: W7221272602