The foundation of artificial intelligence (AI) is algorithmic models and mathematical ideas that facilitate learning and automated decision-making. With an emphasis on supervised, unsupervised, and reinforcement learning paradigms, this work examines the theoretical underpinnings of machine learning and deep learning. The article explores sophisticated deep learning architectures, such as convolutional neural networks (CNNs), for computer vision and recurrent neural networks (RNNs) for sequential data processing. It also discusses autoencoders, which are employed for dimension reduction and anomaly identification, as well as generative adversarial networks (GANs) for the production of synthetic data. To conclude, the article highlights the use of AI in several sectors such as computer vision, natural language processing, healthcare, finance, and additive manufacturing, demonstrating its influence on improving processes and enhancing the performance of intelligent systems.