Artificial intelligence (AI), which has revolutionary potential for applications ranging from telecommunications to quantum computing, is redefining the design and operation of photonic systems. The traditional design of photonic structures relies heavily on physics-based techniques and iterative numerical simulations. Although these techniques have been effective, there have always been limitations in complex design areas, and the ability to reach optimal results has always been a challenge. AI techniques, especially those based on machine learning (ML) and Generative AI (Gen-AI), provide out-of-the-box solutions based on data-driven model simulations that are both forward and reverse. The “real-time” prediction generation of photonic devices and fast search in the design space are some benefits of implementing deep learning techniques. Like reinforcement learning, the design of photonic geometries can be the subject of efficient optimization methods that are enabled by Actor and critic Models. In contrast, PINNs, which are based on physics principles, use Maxwell equations for mode analysis. Due to the design and fabrication of grating couplers and metasurfaces, traditional optimization methods are often outperformed in terms of accuracy and speed. Generative AI oversees changes in photonic computing when photonics-based hardware technologies are developed to allow faster and more energy-efficient neural networks, which in turn can overshadow electrical computing. This combined AI-photonics approach is intended to change the face of things by providing a platform for the development of smart optical systems, holographic imaging, and quantum electronics. However, related issues are numerous. These problems involve quality assurance, data volume, and scalability. The primary solution is to expedite the development of generative AI algorithms and photonics hardware. The future promises photonics as the joining link between the real and digital worlds, and at the same time, the two systems will work together seamlessly, enabling the development of entirely new levels of creative output.