GAN
Generative Adversarial Network — an earlier generative architecture where two networks duel: a generator creates fakes and a discriminator tries to catch them, each making the other better. GANs powered the first wave of AI faces (thispersondoesnotexist.com) and early deepfakes, and they still excel at fast, specialized tasks like face restoration and upscaling. Diffusion models have largely replaced them for open-ended text-to-image because GANs are unstable to train and less prompt-controllable. Knowing the term helps you read AI news and understand where today's creative tools came from.