I don't see this working for Getty but it will be interesting to see. Israel and the EU have preemptively determined that the use of copyrighted data to train M

I don't see this working for Getty but it will be interesting to see. Israel and the EU have preemptively determined that the use of copyrighted data to train ML models is almost always fair use, and I believe the Authors Guild case in the USA also sets strong precedent there. That really leaves the UK, and they are likely to go with everyone else.