I think this is overblown, the watermarking is just enforcing a particular recognizable output text style.
The existing Claudisms like "smoking gun" and "it's not x it's y" evidently haven't impacted the model's agentic capabilities much, and i doubt the new ones will either.
I wonder what watermarks on an LLM output would look like. Say you have a draft for an email you have written and would like the LLM to correct typos and such. Would it be forced to change it substantially? If you tell it to reword it even so slightly, would the watermark still be valid?
3 comments of 8
[ 0.20 ms ] story [ 10.7 ms ] threadThe existing Claudisms like "smoking gun" and "it's not x it's y" evidently haven't impacted the model's agentic capabilities much, and i doubt the new ones will either.