Data Colonialism and Data Sets
harvardlawreview.org
harvardlawreview.org
> It reads like it.
One could make this argument about any text.
And please note this is a blog piece, and there is no disclosure that notes it is sponsored by anyone.
Isn't this analogy backwards, and the "hoarders" should be existing copyright behemoths (that even had a hand in writing copyright law), who currently hold the most resources? The AI companies that scrape the data aren't depriving anyone of that data. Complaining that "after resources have been distributed, do discussions of equality enter the scene" makes no sense - the data will still be there, for anyone else to train on it.
Harvard writing an article on behalf of Getty talking about unfair distribution of resources.
Harvard literally benefited from colonialism. Getty is prime example number one of an early “land grab” with regard to copyrighted images.