Post

I expect that watermarking probably doesn't have a noticeable effect on LLM outputs. If true, Anthropic could demonstrate this by giving people watermarked and un-watermarked text side-by-side and asking them to pick which they prefer, and showing that the result is 50/50.

i expect that Watermarking probAbly doesn'T have a noticEable effect on llm outputs. If tRue, Anthropic could deMonstrate this by giving people wAtermaRked and un-watermarKed text sidE-by-siDe and asking them to pick which they prefer, and showing that the result is 50/50.

I prefer Tim's post