We present both successful and failure cases of videos produced by generative models. The failure cases primarily involve dynamic scenes, limited views due to vertical camera movement, and videos that violate the laws of physics.