AI-enabled disciplinary construction and assessment: dilemmas and pathways

AI has shown significant potential in enhancing the construction and high-quality assessment of academic disciplines. As China transitions from an index-driven to a high-quality development model in disciplinary construction, the focus has shifted from using evaluation indicators to solidify disciplinary foundations, to addressing societal concerns and improving overall standards through assessment, and ultimately toward stimulating the internal motivation for excellence. However, during this critical transformation period, AI-enabled disciplinary construction and evaluation still confront three major challenges, namely ideological inertia, technological limitations, and a lack of ecosystem support. Meanwhile, the historic push toward building a strong nation has accelerated the need for innovation in talent cultivation models, research paradigms, and evaluation mechanisms. In response, this study proposes a strategic path forward through deepening conceptual understanding and breaking cognitive barriers to integrate AI more effectively into decision-making process, building collaborative, open, and value-driven evaluation platforms to foster multi-stakeholder engagement, and advancing human-AI collaboration to drive adaptive transformation in assessment organizations through technological innovation, thereby creating a governance ecosystem characterized by human-machine complementarity and flexible responsiveness.