In recent years, there has been a surge of interest in creating virtual models of human cells using artificial intelligence (AI) and machine learning techniques. These virtual cells can simulate the behavior cells based on multi-omics data, which is essential for understanding cellular processes and disease mechanisms. However, current AI models are black boxes, making it difficult to interpret their predictions. In this article, I give my perspective on the challenges of using AI for modeling human cells and the importance of developing interpretable models that can provide insights into cellular mechanisms.