Anyone who has written anything would, I hope, understand that the difference between the sentences “The weather today was cold and grey” and “The weather today was cold and overcast” are sometimes “low stakes,” as Anthropic describes, but not always. “Grey,” and “overcast” are different words, and there are any number of reasons why a human author might pick one over the other in a given context.

This isn’t an issue with their watermarking process, it’s an issue with LLMs as a whole. LLMs are built entirely out of probabilities rather than any sort of logic or intent.