This paper presents an empirical analysis of the effectiveness of static analysis and artificial intelligence-based analysis in detecting plagiarism in source code. The study describes a custom solution for static analysis, as well as the selection of the large language model DeepSeek in the cloud for plagiarism detection purposes. The paper emphasizes the importance of developing tools for plagiarism verification. Furthermore, the advantages and limitations of each analysis method are highlighted depending on the type of modifications in the code. Based on the obtained results, directions for further development are proposed, including a hybrid approach that combines both methods.