The data is already incorporated into existing models. Those models will generate derivative data used to train the next models. They only needed to rip it all off once, after that it's actually preferable to not let anyone do it again.
No comments yet.