Generative Adversarial Networks (GANs) are a type of artificial intelligence model used to generate new data that resembles existing data. They consist of two neural networks: a generator that creates new samples and a discriminator that evaluates them. The generator tries to produce data that is indistinguishable from real data, while the discriminator attempts to differentiate between real and generated data. Through a process of competition, both networks improve over time. The generator learns to create more realistic samples, while the discriminator becomes better at identifying fakes. This adversarial training leads to high-quality outputs, making GANs popular in fields like image generation and video synthesis.