# GANs **Generative Adversarial Networks (GANs)** are a training framework where two neural networks compete: a *generator* that learns to produce convincing fake data, and a *discriminator* that learns to tell fake from real. Each one makes the other better — the generator improves by fooling the discriminator, and the discriminator improves by catching it. When training succeeds, the generator can produce outputs indistinguishable from real data. ^overview > [!example] > Think of a counterfeiter (generator) and a detective (discriminator). The counterfeiter gets better at forgery, the detective gets better at spotting fakes — until the forgeries are flawless.