Iris biometric authorization enables seamless authentication, preventing the spread of infectious illnesses like COVID-19. Due to its intricate texture and distinctive traits, the iris is often analyzed to recognize and identify an individual across most applications. Even though the iris has a distinct textural design, it might be replicated. So, this work aims to achieve robust iris-liveness detection against adversarial attacks. It proposed a unique ILD system that uses a GAN-based spoofing assault generation technique. The Efficient NetB7 is used to categorize spoofing attacks into distinct categories since it is necessary to recognize various types of adversarial iris spoofing attacks. The study addressed the critical problem of detecting adversarial iris spoofing attacks using DCGAN and Efficient NetB7.The novelty of this study is that it identified a total of seventeen iris spoofing attacks using DCGAN and Efficient NetB7 with reasonable accuracy. This proposed work identified known and Adversarial Iris spoofing attacks with 99% accuracy. GAN-based transfer learning offers enormous promise in a related field and may be investigated further depending on the research needs.