We study object detection and image classification models and observe that their respective architectures are vulnerable to image distortions such as noise, compression, blur or snow. We propose alleviating this problem by training the models with antibodies generated using Artificial Immune Systems (AIS) from original training samples (antigens). These antibodies are AIS-distorted antigens at the pixel level through cycles of “select, clone, mutate, select” until an affinity to the antigen is achieved. We then add the antibodies to the antigens, train the models, validate and test them under 15 distortions, and show that our data augmentation approach (AISbod) significantly improved their accuracy without altering their architecture or inference speed. For example, YOLOv4 improves by 3.90% on average over all 15 distortions, 4.06% under snow, and 28.11%  under impulse noise. Our simulations show that compared to related defence methods, our method performs better under distortions, is more consistent across datasets and object detection models, and also under clean samples. Moreover, we show that our approach to image classification significantly improves accuracy under distortions.