The article keeps making it sound as if it's a problem for humans. e.g.: > Now here the date is more flexible, let's say 2022. But if you're collecting data be

The article keeps making it sound as if it's a problem for humans. e.g.: > Now here the date is more flexible, let's say 2022. But if you're collecting data before 2022 you're fairly confident that it has minimal, if any, contamination from generative AI. Everything before the date is 'safe, fine, clean,' everything after that is 'dirty.'" Though what it seems to actually mean is that it's a problem for (future) generative AI (the "genAI collapse"). To which I say;