Cosmic rays (CR) reaching telescope detectors in outer space are known to induce glitches and background noise. The presence of CR noise significantly influenced Cosmic Microwave Background (CMB) experiments, like Planck and LiteBIRD, which have a long exposure and hard shelling or filtering. In order to address this challenge, it is imperative to accurately simulate the CR background throughout the duration of LiteBIRD’s three-year mission. However, state-of-the-art Monte Carlo (MC) simulations are extremely computationally expensive, typically requiring 30 times the simulated period. We present the Cosmic Rays Artificial Background (CRAB) code, extending MC simulations with Generative Adversarial Networks (GAN). By leveraging GANs, we can efficiently generate a sufficient number of genuine, statistically independent images, unlike traditional noise analysis techniques combined with template expansion methods.