We are missing a big opportunity here! Training datasets should be considered a worldwide Commons and any profits made from those training sets should be fed ba

We are missing a big opportunity here! Training datasets should be considered a worldwide Commons and any profits made from those training sets should be fed back into national (or even worldwide) sovereign funds. If they go to profit from our work, we should get a share of those profits equivalent to or in excess of what shareholders make.