Videos obtained by super-resolution techniques are getting closer and closer to real high-resolution videos, but the authenticity of these videos is unknown, and it is impossible to tell whether the resolution of the video is real or a fake video obtained from a low-resolution super-resolution. This creates difficulties for tasks that require very strict video authenticity. Deep learning models are particularly good at processing large amounts of visual information and can capture subtle differences that are hard for the eye to observe. Therefore, the high-resolution features of videos can be effectively analyzed and recognized using deep learning techniques to distinguish the authenticity of video resolution. In this paper, we propose a way to determine the authenticity of video resolution using a super-resolution generative adversarial network discriminator. By training the model, a high-performance discriminator is obtained for discriminating between real high-resolution videos and super-resolution generated fake videos. Results of the experiments show that the approach is capable of detecting the resolution authenticity of an image or video and provides a feasible solution for video authenticity detection efforts.