Artificial intelligence (AI) is rapidly transforming atmospheric prediction by offering faster, data-driven approaches that complement traditional physics-based methods. This article examines the rise of modern AI climate models, with a focus on foundation systems such as Microsoft’s Aurora, alongside GraphCast and Pangu-Weather. It highlights how these architectures leverage pattern recognition and latent representations to deliver high-resolution forecasts, while also exploring emerging directions like hybrid modeling and agentic AI. Key challenges, including data bias, interpretability, and the environmental cost of large-scale training, underscore the importance of responsible and sustainable deployment. By blending technical insight with practical context, the article aims to help students understand how AI is reshaping climate and atmospheric science and what this evolution means for future research and engineering careers.