@pcdevil@mastodon.social Why? Isn’t this what LLMs are actually good at: being trained on large corpuses of stuff to perform statistical analysis on similar/dissimilar stuff? What _are_ we supposed to use to detect AI-generated media?