Chat-GPT does not understand any of the following:
- engagement
- human tone
- readability
- conversational tone
- humor
- transitions
- or basically anything to do with quality content
I keep training it on what to do instead of what not to do, even down to the word and phrase level in custom GPTs, and it STILL forgets things or messes things up.
It does decently well in one paragraph with specific instructions but then forgets things and reverts back to its very, very poor tone of writing.
It’s so easy to spot Chat-GPT content in a microsecond. Is anyone in the company working on improving the tone of it?
I can point out 50-100 fixes that need work immediately.