Abstract: Foundation pit engineering is an important branch of engineering. With the rapid global advancement of artificial intelligence (AI) technology, its applications in foundation pit engineering have extended to areas such as deformation prediction, safety assessment, and support design optimization. This paper systematically explores the application pathways of AI in scenarios like deformation prediction, image recognition, and intelligent monitoring by comprehensively employing machine learning and deep learning algorithms including support vector machines (SVM), time series analysis (ARIMA), and convolutional neural networks (CNN), combined with multi-source data fusion and comparative analysis methods.The research findings indicate that AI technology can effectively reduce deformation prediction errors and construction costs, while achieving millimeter-level displacement early warning through multi-source data fusion. However, the study identifies data quality, model interpretability, and shortages of interdisciplinary talents as major obstacles to technological implementation. Future research should focus on the development of interpretable AI models, the construction of multi-source data knowledge graphs, and innovations in interdisciplinary education systems to promote the intelligent upgrading of foundation pit engineering.