You know, all of a sudden, I am starting to lose interest in meticulously drawn Mermaid diagrams in README, perfect grammar and spelling in doc reviews, or neat generated general photographs. They are all correctly presented, of course. But the ideas are mostly wrong, too.

I guess my stick figure hand drawn diagrams, a doc with few mistakes in grammar or spelling would be seen as more worthy to read as long as my ideas are sound. Right? :-)

Yes, genuineness, authenticity, quirky imperfection will be prized. But presumably some of that can also be trained into the models so...

If this becomes a trust signal, you can prepare for next gen models to do stick figure hand-drawn-like diagrams with spelling mistakes.