If Claude makes the third paragraph clearer, whose work is it?

Anthropic are now watermarking everything Claude writes.

I personally don't hate it, but it does put a question underneath everything now.

The model quietly biases its word choices so a pattern sits underneath the writing, readable by a tool, invisible to you. Other labs signed the same code of practice with the EU, so this will eventually be across all models.

But if I have the idea, build the argument, make the judgement calls, and mean the thing I'm saying, and Claude makes the third paragraph clearer, whose work is it?

If I wrote the idea, the argument, the meaning, the emotional core, and the final judgment, then it is my work. If Claude helped translate it, polish a sentence, or make it clearer, that does not suddenly make Claude the author.

But a watermark probably changes how people perceive authorship. People are not going to say, "Claude may have been involved in processing this text." They are going to say, "AI wrote this."

We spent years learning that a metric is not the thing it measures. Impressions don't equal attention. Clicks aren't interest. Now we've built a signal for machine involvement, and I think many may start reading it as a verdict on human effort.

It marks the machine's involvement. It can't mark the human's.