Created from a single voice note with Agent Craft

I looked at the blog post they put out yesterday on the Claude…

I looked at the blog post they put out yesterday on the Claude watermark and it is quite clever. They are using a Google technique. If you have a sentence with a word that could be swapped without changing the meaning, they generate that word with a random number. Now instead they use a different random number. There is a method to how they pick that additional word and that is what can be traced. The net impact is that it should not change the meaning of your content. Let me dispel a couple of myths. It is not going to cost more data center processing so it is not bad for the environment. It is not going to cost a lot of processing to detect either. The detection mechanism is designed to be lightweight, and even if it is used extensively it is not going to cost a lot. So there is no environmental concern. The second concern is people said this is a copyright play. They are going to claim copyright over all of the content that Claude has used. But as far as today if you look at their terms and services there is no claim on copyright although that could change in the future. We need to keep an eye on that space. It does look as if they are doing this just to meet European Union regulations which is kind of annoying. There is no detrimental cost to people. Now here is where it gets interesting. When other platforms like LinkedIn or Facebook want to detect whether the content you are writing and submitting has been produced by Claude or not they are now going to be able to easily do so. And this is very likely to affect your reach.  I have to admit that I do have a bit of shadenfraud from all these people that were complaining about AI slop and now they’re complaining about the ability to detect AI slop in their own writing. Building: AgentCraft (dot) ai The app that grows your personal brand in minutes per day.