In the era of big data, artificial intelligence (AI) technology has developed rapidly, with medical field being one of its most deeply penetrated application domains. Leveraging the advantages of big data and AI requires the sharing and integration of medical data, yet balancing privacy protection and data sharing poses significant challenges. This paper analyzes the focus issues of patient privacy and data security in medical AI from a global perspective across 6 dimensions: challenges in data sovereignty and cross-border flow compliance, technical vulnerabilities in de-anonymization and re-identification risks, failure of informed consent mechanisms and dynamic authorization needs, regulatory gaps in algorithmic “black boxes” and data misuse, technological dependency and supply chain security risks, and the dilemma of balancing privacy protection with public health interests. Corresponding solutions and strategies are also proposed.