Ironically this reads like AI slop.

Big LinkedIn post on a concept with little proof.

> If I subconsciously detect that you spent 12 seconds creating this, why should I invest five minutes reading it?

The problem is it isn't easy to detect it and I'm sure the people who work on generated stuff will work hard to make detection even harder.

I have difficulty detecting even fake videos. How can I possibly I detect generated text in plain text accurately? I mean I will make plenty of false positive mistakes, accusing people of using generated text when they wrote it themselves. This will cause unnecessary friction which I don't know how to prevent.