<![CDATA[This research examines how artificial intelligence (AI) technology could potentially address the rural-urban education gap in China through a qualitative study of its application in under-resourced rural schools. The research adopts a mixed-method design, comprising a policy analysis of China's 2024-2025 AI in education guidelines, case studies of three rural primary schools in Guangdong and Guizhou provinces, and semi-structured interviews with 10 key stakeholders, aimed at assessing the impact of AI tools and customized learning platforms on issues relating to inadequate staffing, rote teaching, and the absence of individualized instruction. Findings demonstrate that AI technologies can increase learner engagement, allow for the early identification of learning gaps, and assist in providing necessary instructional support. However, persistent structural barriers, including uneven digital literacy, deficient infrastructure, and inadequate teacher preparedness, alongside ethical concerns over data use and algorithmic bias, pose fundamental challenges to equitable implementation. The study reveals that robust investment in rural digital infrastructure and sustained, context-sensitive teacher training are prerequisites for realizing AI's transformative potential. Crucially, it argues that without targeted structural interventions, AI risks exacerbating existing educational inequalities rather than mitigating them. The primary contribution of this study is its identification of an implementation gap between national policy and local reality, and its proposal of a bottom-up, co-construction framework. This framework, derived from the empirical experiences of rural stakeholders, offers policymakers and educators a pathway to harness technology more effectively for improving learning opportunities in China's marginalized rural areas.]]>
