The adoption of artificial intelligence (AI) is increasingly seen as a pivotal step in modernizing internal audit systems within universities, steering them toward a more intelligent and adaptive future. By embedding AI into audit processes, institutions can significantly sharpen their ability to spot risks and enhance the level of audit decision-making, ultimately raising the bar on audit quality, operational efficiency, and supervisory rigor. This paper puts forward a structured framework to guide the intelligent transformation of university internal audit, organized around two core dimensions: business logic and technical implementation. On the business side, AI enables the streamlining of audit workflows and the smarter configuration of audit content, setting the stage for a seamless shift into the digital-intelligent era. Technically, it allows for end-to-end intelligent coordination from data collection and cleansing, through modeling and analysis, to the automated generation of audit reports, paving the way toward the construction of intelligent audit agents. The study identifies potential challenges in this transformation process, including data security risks, dilemmas in algorithmic interpretability, and inefficiencies in human-machine collaboration. In response, the paper proposes a systematic approach to overcoming data silos and improving governance structures through embedding interpretable models in compliance frameworks, establishing manual review mechanisms, forming interdisciplinary teams, and enhancing auditor competence training. These measures are essential to advancing the transition of internal audit in universities from informatization to intelligentization.