肾上腺肿瘤是具有高度异质性的内分泌系统疾病，其临床亚型间预后差异显著，因此精准诊疗至关重要。为提升诊疗精度、优化预后评估、推进个性化医疗，人工智能（AI）在肾上腺肿瘤领域的应用日益广泛。本文系统总结了近年来AI在肾上腺肿瘤精准诊断、治疗策略制定和预后预测等方面的应用，并探讨其优势与局限。AI可有效提高肾上腺肿瘤精准诊疗水平，在诊断层面，实现亚型鉴别受试者工作特征（ROC）曲线下面积（AUC）达94.5%；在术中管理上，血流动力学风险预测AUC达93.7%，在预后评估中，对肿瘤转移的预测敏感性达91.7%，特异性达95.5%。此外，AI还可有效辅助手术决策与个体化药物推荐。然而，当前应用仍面临诸多挑战，包括单中心数据异质性高、算法可解释性不足、多模态与多组学融合不充分等。未来需通过构建标准化多中心数据库、发展可解释AI技术、强化多维度信息融合及开展大规模临床验证等措施，加快AI在肾上腺肿瘤临床中的转化，推动肾上腺肿瘤诊疗向精准化与个体化发展。

Adrenal tumors are highly heterogeneous endocrine disorders with markedly varied prognoses across clinical subtypes, underscoring the critical need for precise diagnosis and treatment. Artificial intelligence (AI) is increasingly employed in this field to improve diagnostic accuracy, refine prognostic assessment, and advance personalized medicine. This review systematically summarizes recent advancements in AI for the precise diagnosis, treatment planning, and prognosis prediction of adrenal tumors, while examining its strengths and limitations. AI has significantly enhanced precision in managing adrenal tumors. It achieves an area under the curve (AUC) of 94.5% for subtype classification, 93.7% for predicting intraoperative hemodynamic risk, and 91.7% sensitivity for tumor metastasis, and 95.5% specificity for prognosis assessment. Additionally, AI effectively supports surgical decision-making and personalized medication recommendations. However, current applications face several challenges, including high heterogeneity in single-center data, insufficient algorithm interpretability, and inadequate integration of multimodal and multi-omics data. Future efforts should focus on establishing standardized multicenter databases, developing explainable AI techniques, enhancing multidimensional data integration, and conducting large-scale clinical validation. These measures are essential to accelerate the clinical translation of AI and to propel the field toward more precise and individualized patient management.