I think there's a lot of behind the scenes manipulation of the model to steer clear of anything that would bring negative criticism of its output.
Surely that leaks into what lyrics it chooses to write, what stories it can produce.
Consider https://platform.openai.com/examples/default-marv-sarcastic-... as one of their standard examples.
ChatGPT has guardrails as its easily accessible and people are using it... and posting what they get it to say.
It is much less compelling to say "here's a curl command that I invoked that got this response back."
[1]: https://www.theguardian.com/us-news/2023/feb/06/nothing-fore...
I think there's a lot of behind the scenes manipulation of the model to steer clear of anything that would bring negative criticism of its output.
Surely that leaks into what lyrics it chooses to write, what stories it can produce.