Neural network chip technology is a branch of AI (Artificial Intelligence) that uses large-scale ICs (Integrated Circuits) to simulate the neural patterns of the human brain in a systematic way. The technology can guide computers to process data in a way that is similar to a human brain. The neural pattern system is modeled after how the biological neural system is made up, how signals are transmitted, and how it processes and stores information. This involves electronic circuit materials, components, circuit simulation, circuit design, computing architecture, algorithms, and system engineering simulation. This report focuses on three areas: types of AI chips, characteristics of neural network chips, and strategies of major brands.