Abstract: With the deepening advancement of rural revitalization strategies,AIGC (artificial intelligence generated content),with its rapid response capabilities and diverse content generation,precisely meets the dynamic and differentiated needs of rural construction.It provides core support for preserving rural collective memory and cultural genes,shaping livable environments,and enhancing rural attractiveness.This study,taking Zengcheng district in Guangzhou as an empirical case,proposes a closed-loop technical framework of “feature cognition-feature learning-feature generation.”It systematically implements geo-cultural spatial gene decoding, construction of a multimodal dataset for vernacular architecture,and multi-scale scenario validation of AIGC generative models.The research explores synergistic mechanisms for AIGC-driven targeted generation of distinctive rural landscapes.Key outcomes include:①Achieved controllable landscape generation under cultural-geographical constraints; ②Established a three-tier advancement paradigm (data-algorithm-application); ③ Provided technical methodological support for rural cultural preservation and spatial governance in the AI era.

Key words: rural revitalization, AIGC, geocultural space, rural landscape, generative model, artificial intelligence