To explore the value positioning of artificial intelligence (AI) in basic education

this paper reconstructs an educational framework centered on developing higher-order thinking based on the intrinsic logic of human learning. Through theoretical deliberation and literature analysis

it elucidates three fundamental characteristics of human learning: the understanding of the essence of tacit knowledge

the attention-driven mechanism of deep learning

and the inherent contextuality and embodied nature of cognition. It further argues that AI-driven transformations in knowledge production necessitate shifting foundational education goals from knowledge acquisition to cultivating higher-order competencies. Analysis reveals that current AI applications often deviate from the intrinsic demands of deep learning. Consequently

this paper redefines AI-supported foundational education pathways: constructing embodied

contextualized intelligent environments; emphasizing the foundational value of basic knowledge in supporting higher-order thinking; establishing learning mechanisms where knowledge and skills mutually construct and coexist; and designing human-centered auxiliary systems with information filtering capabilities.