Hacker News Reader

For text, have a big model generate the "intelligent" answer, and then ask a local LLM to rephrase.

Yeah exactly, you can always do that by using another model that doesn't have the watermark.