There is a storm coming for artificial intelligence relating to the use of copyrighted materials for training. Just because something is “publicly accessible” doesn’t make it OK to use in deep learning model-training and it might be time for some deep-unlearning.
