Insights on this cat-and-mouse game are spot on. As AI evolves, watermarking might be a viable countermeasure. what detection methods/tools have you found most reliable in practice? I want to be able to detect and automatically block AI generated comments under my posts.