Copyright is long, but not infinite. There's little point in my opinion to permissions to train upon unless your purpose is to simply slow and postpone the inev

Copyright is long, but not infinite. There's little point in my opinion to permissions to train upon unless your purpose is to simply slow and postpone the inevitable. I'd be more interested in getting AI companies training with publicly scraped datasets to have them available to other AI trainers for free. The load scrapers put on the web is out of control.