I’d love to see the actual prompts you provided to ChatGPT. And which model (3.5 or 4) that you were using. I ask, because I’m not able to replicate your results.
3.5.
And the prompts were to get it to critique the Bible or Koran at all, then tell it to be more and more positive and effusive (or negative and critical). After a few iterations, you’ll get something similar to mine.