Which component tries to generate fake data that the discriminator cannot distinguish from real data?

• A. Generator in GAN
• B. Discriminator in GAN
• C. Training Loop
• D. Prompt Engineering

Answer: (A)

In a Generative Adversarial Network, two networks compete: a generator and a discriminator. The generator creates synthetic data from random input, aiming to resemble real data. Its goal is to produce samples that the discriminator cannot reliably distinguish from real ones. The discriminator, on the other hand, evaluates inputs and tries to tell real data apart from generated data. The generator is the component that generates fake data to fool the discriminator, which is why it’s the correct choice. The discriminator’s role is the opposite—detect fake data—while the training loop coordinates the alternating training, and prompt engineering is unrelated to this GAN interaction.