With the rapid development of artificial intelligence (AI) technology, especially deep learning, AI is playing an increasingly important role in the diagnosis and treatment of liver cancer. AI has shown great potential in improving diagnostic accuracy, reducing the workload of physicians, and providing personalized treatment plans. In imaging diagnosis of liver cancer, AI technology has been widely applied to ultrasound, CT, and MRI data, enabling automatic recognition and segmentation of liver lesions and differentiation between benign and malignant lesions. In some cases, its diagnostic accuracy can match or even surpass that of professional radiologists. In pathological diagnosis, AI assists pathologists in differential diagnosis and classification by analyzing whole slide image, improving diagnostic efficiency and accuracy. For treatment decision support, AI analyzes clinical, imaging, and pathological data to predict postoperative recurrence risk, evaluate the effectiveness of specific treatments, and identify biomarkers, providing new perspectives for precision medicine and assisting clinicians in designing personalized treatment plans. With the advancement of multimodal AI models that integrate imaging, pathology, and clinical data, it is possible to achieve a more comprehensive understanding of the biological behavior of liver cancer, thereby offering more precise diagnoses and treatments for patients. Despite its immense potential, AI in liver cancer diagnosis and treatment faces challenges such as data standardization, security, and privacy protection. The clinical translation of AI models requires further validation and refinement. With technological advancements and data accumulation, AI is expected to provide more precise and personalized medical services for liver cancer patients in the future. This article focuses on summarizing significant research advances in AI for liver cancer diagnosis and treatment, and briefly discusses its limitations and future directions.