> I recently made a user script that tries to highlight LLM generated text
How does that work?
How does that work?
It has a pretty high false positive rate though, but it reliably highlights AI generated spam websites and saves me from having to read them.